CA2654866A1 - System and method for auctioning avails - Google Patents

Abstract

A system and method is provided for use in connection with auctioning delivery spots (e.g., ad spots) or commercial impressions in a broadcast network. The system provides (1702) information regarding asset delivery spots and receives (1704) bids from asset providers. A winning bidder is determined (1706), and a corresponding asset is delivered (1708) via the network.

Description

SYSTEM AND METHOD FOR AT.~CTIC3NI~'~G AVAILS
CRI~~~REFEIIENCE
This application claims priority under 35 U.S.C. 119 to U.S. Provisional Appllcation'No. 60/804,459, entitled: "ADVATAR AI~41v3 AUCTIONS," fileal on June 122 2006, the contents of which are incorporated laereln. as if set forth in full.

FIELD
Systems and methods presented hc-regn relate to the provision of targeted assets via a network interface. In one specific a~rangenieii:t, targeted ~dvetlslng media delivery spots are aucta~~~ed to asset providers (e.g., advertisers).

BACKGROUND
Broadcast ~~twork-. content or programming is conirnanly provided in ~onjulictlon with associated iiifarmational content or assets, These assets include advertisements, associated programming, publicmservl~e announcements, ad tags, trailers, weather or ~~~~~~ency notifications and a variety of other content, including paid and unpaid content.
In this regard, assets providers (e.g., advertisers) who wish to convey lnI'orinatgon (e.g., advertisements) regarding services and/or products to users of the broadcast network often. pay for the right to insert their ifl~formatlon into programming of the broadcast network. For instance, advertisers may provide ad content to a network operator such tl}at the ad content may be interleaved wlth broadcast network programming during one or more programmlng breaks, 'I"h.e delivery of such paid assets of1:en subsidizes or covers the costs of the progranuni.ng provided by the broadcast network, This may reduce or eliminate costs ~~~e by the users of the broadcast network lsrc~gTamming.
In order to achieve a better return on their investment, asset providers often try to target their assets to a selected audience that is believed to be interested in the goods or services of the asset ~rovlder. '1h~ case of advertisers on a cable television network is illustrative. For instance, an advertiser or a cable television network mav desire to target its ads to certain derr~ogTaphl~ groups based on, for ~xamp1e, geographic locationq gender, age, income etc. Accordingly, once an advertiser has created an ad that is targeted to a desired group of viewers (e.g., targeted group) the ~dvertis~.~ may attempt to procure I

insertion times in the network programming when the targeted. gToup is expected to be among the audience of the network programming.

SUMMARY
The inventors of the present application have recognized that systems that allow for obtaining information regarding current network users and./or the ability to d~iamically insert assets (e.g., ad content) into one or more content streams may allow asset providers to more effectively match their assets to targeted network users. Ftirther, the inventors have recognized that the ability to, inter alia, ~~tai~~ current itiformation and/or siynainically ins~it assets into one or more content streams of a ba~~~a(i.cast network.
may facilitate additional i'-Ãmctionalities for targeted advertising. For instance, in one aspect, functionality is provided for auctioning asset delivery spots (e.g., avails), an audience segment of an avail, or an aggregated audience (e.g., aggregated.
across multiple avails and,%~r multiple bandwidth s~~ents or channels), to asset provide s.
Sucb.
auctioning may be done interactively prior to specific avails and/or in an automated process.
According to a first aspect of the present invention, a method and apparatus (utility) is provided for use in auctioning specific avails in a broadcast content stream.
Generally the utility is utilized in connection with delivering assets to users of a broadcast network that is pn'marily utilized for symch~oni~ed distribution of broadcast content to multiple users. The utility involves providing zn~~rinati~~ regarding an available asset delivery spot. Such infoarnation may be provided to a plurality of asset providers. Bids are then received for the asset delivery spot from one or more asset providers. Based upon a winning bid, an as-s~t of one of the providers (e.g., the winning bidder) may be inserted into a content stream of the broadcast r~~nvork for delivery during the asset delivery spot. Finally, a payment associated with the winning bid may be adjusted based upon infannatis~~ regarding the delivery of the asset to the users of the broadcast network.
Such adjustinen~ ~~av be based upon. the actual deliveries (e.g., total number of receiving users) and/or based upon statistical inf~~ation associated W itli the estimated delivery of the asset to the users of the broadcast nehvrsrk.
'~e assets may include any assets of an asset provider, including, without limitation, advertising, programming, public service announcements, etc.
Further, various steps and processes may be perf'onnetl at different locations within a broadcast network.

A number of these steps and processes may be performed at a location that is referred to as the headend of the network, which may generally entail any nehvork components that are upstream (i.e., across a network interface) of customer premises equipment (CPE) or other user equi~inent devices (UE.Ds) (e.g., including mobile devices of a subscriber sucb as mobile televisions, data enabled phones and hard drive devise). Likewise, asset providers may be able to access ~iie or more components of tlie system across one or more network- interfaces for atictioni~~ purposes.
Infonnation regarding art, available asset delivery spot may include temporal in~~nnation regardiiig the asset delivery spot andEor network c1~aimel infonnation for the asset delivery spot. hi this regard, a time, location and'or program associated with the asset delivery spot may be provided. In ~further arrangement, the lni`orxnaxi~~~ inay include ratings inf~nnatiogi for programming associated with the asset delivery spoI.
Such ratings may include overall ratings for the programming and/or ratings that are arranged by one or more characteristics of network users that receive the programming, For instance, ratings may be provided that are arranged by different d~~graphic groups of network users. In addition to providing ratings information, audience classiffil cation parameters may be provided with the asset delivery spot. For instance, such audience classification parameters may relate to age, gender, income, geogTaph.ic local, and1~~
audience iiiterest. In any case, asset providers may receive sucli in~~rination for use in d~~enniring the desirability oI'bidding on the asset delivery spot.
Receiving bids from the asset proNriders may include i~~~eractively receivitig bids during a predetermined auctioiiiiig pe i.od in which asset providers may bid competitively with one another. In a fiarth~r arTanI;ernenI, receiving bids includes receiving predefined ad ~~~paigns for one or more asset providers where such ad campaigns include one or more targeting criteria and bid infÃ~rmation. In such an arrangement, the utility can correlate xl~e targeting criteria and bid inI`onnation of the predefined ad cair~paigns with information regardigl,~ an available asset delivery spot to identify a wining bid. ~ongst a plurality of predefined ad campaigns. In this later arrangement, such bidding may be an automated process that is based on the predetermined criteria of various asset providers.
The automation of the auction process may allow for performing auctions a shorl: time prior to the asset delivery spot. For instance, such an auction may be performed during the p~ograin in which the asset delivery spot is included. In this regard, the system may be an on-the-fly oxj ust-in-ti~e auctioning system.

Inserting an asset of the winning asset provider may include replacing a default asset in the content delivery stream with the asset (i.e., winning asset) associated with the winning bidder. In another wTangernent, inserting may include transmitting instructions to at least a portion of U:EDs within the broadcast network to play the winning asset during the asset delivery spot. Iri such an arrangement, the winning asset may have been previously stored in storage facilities of the '~T EDs. In another arrangement, inserting may include inserting the asset associated with the winning bid into a parallel ~ontent. stream 1:hat is broadcast in synchrony witli a content s~~~n that includes the asset delivery spot.
In this regard, lnI'~~ination may be provided to at least a portion of the UEDs in the IS~~~dcast network regarding the availability of the asset on the parallel content stream.
Such ln~onnatioii provided to the UEDs may include one or more targeting criteria for the asseto The UEDs can Ihert, switch to the asset on the parallel content s~~ani during the asset delivery spot.
As will be further appreciated, utilization of parallel content streams ~iav allow for insertir~c, a second, third or more assets of additional asset. providers (e.g., additional winning asset providers) into additional parallel content st~earns.
Accordingly, the UEDs of the network users may be operative to select an appropriate asset for individual network users (e.g., based on a user profile stored by each IJED). In this case, the different parallel asset insertion spots of the parallel content streams may be individually auctioned. For instance, such individual auctions may be based on different demograplxics.

.In order to adjust the IsayTnent associated with the winning bid, it may be useful to obtain information associated with the actual delivery of the asset of the winning asset provider(s). This may entail obtaining statistical gnfc~rrnation associated with the delivery oI't~e asset to the network users. In another ~ang~ni~iit, users ~~~t actually receive the asset may provide a notification to the utility in order to determine an ~ctLial nLimber of users who received the asset. In any case, the bid a.rnÃ~~t associated with.
tlie winning bid may be adjusted based on the delivery of the asset to r~~hvork users. For instance, if the asset is delivered to a number of network users that is less then expected, the payment associated with the wlmling bid may be reduced, and ~~~e versa.
The bidding mechanism may be based on a highest bid aaT~~~ement. 'I'hat is, the bid that results in the greatest payment for the network provider may be utilized.
However, it will be appreciated that any ot1~er applicable bidding system may be utilized.

In tMs regard, a. Vickery bidding system may be gmplt-mented in which the winning bidder pays a nominal (e.g., one penny) amount more than the second place bidder.
In accordance with another aspect of the present invention, a utility is provided that allows for auctioning multiple available asset delivery spots for a common avail. For example, a parallel content s~ewn(s) may be utilized to deliver at least first and second different assets to first and secoiiÃ1 different sets of viewers of a common broadcast progrwno In this regard, the utility involves auctioning an asset delivery spot of a broadcast program that is to ~~ broadcast to network audience to two or more asset providers (or potentially to a single asset provider). Typically, the auctioning is based on at least first and second audience c1~~~~~teristacs (e.g., cjl~~~~t demographics). Once completed, first and second assets associated wixli the winning bids are inserted in to one or more content streams of the broadcast x~~hvorlÃ. Accordingly, dunng the broadcast of the prograrn the first asset may be delivered to a first portion of the network audience a:aid the second asset may be delivered to a second portion of the network audience during the asset delivery spot. In this regard, the f-irst and second assets may be delivered simultaneously. In such an aiTangÃ;rnent, the first a-nd second assets may be delivered oil separate broadcast content s~earais.
The characteristics of the audience may be any characteristics that may be readily defined. Such characteristics may include demographics and/Ã~~ ratings. For instance, an auction based on ratings may allow for first and second asset providers to each provide their assets to, for example onemlÃalf of the audience. In this regard, the audience characteristic is a percentage of the total audience. More cÃsmnionly, such characteristics ma~r be based on ~is~as~gTaplics to allow for more specific targeting of members of the broadcast audience. 'l"l~~ process of auctioning may i~ivolv~ ~~orinflng a first auction for a first audience c1~~~~~~erastac and perfiomi1tig a second auction 11~r a second audience characteristic. In conjunction with auctioning, inforinatior~ may be provided regarding the asset delivery spot. Sucl~ inforinatlÃrn may lncluÃ1.e, without limitation, ratings anÃl,lÃ~r demographic in~ormatioai.
Delivering the first and second assets may include delivering a first asset within a content stream including the broadcast prr~grarn delivering the second asset on a parallel content delivery stream. Altematively, both assets may be provided in parallel content streams. In any arrangement that utilizes ~aralle-l content strearnsy information mav be provided to ~'EDs within the network to allow the UEDs to change between content streams based on, for exainple, audience classification parameters d~tenni~ed at the UEDsa Altemativelyj the UEDs may randomly select between available assets for delivery during the asset delivery spot. For ~xainples in instances were first and second asset providers have won one-half of the available network audience, the UEDs may be operative to select either or the first or second asset during the asset delivery spot. In any case, the UEDs may be operative to provide tioxificatior~ of the receipt and/~r output of the first and/or second assets.
According to a furl~er aspect of the preseiit invention, a utility is provided for auctioning a~idience impressioiis. In this arrangement, audience iinpressions (e.g.5 delivery of an asset to a single household or ~etWork- user of a household) may be auctioned to asset providers. The utility involves offering a ~~edeterÃni~ed number of audience impressions for an audience having at least one predefined targeting crite-i.a.
For example, a million iinpressl~ns may be provided for a specific demographic, e.g.}
males between the ages of 18 atid 24. The number of audience iinpressions may be provided witliin a set tirne period (e.g., 24 hours, 48 hours, etc.).
Accordingly, the utility may receive bids for the audience impressions from one or more asset providers.
Likewise, the asset of a winning asset provider may be delivered to network users having the at least one defined targeting criteria. Suc-h targeting criteria may relate to ~~~~gapbic infonnation such as age, gender, income, geographical location, etc.
Such delivery may be perfonned in a single asset delivery spot. In another aaTan,~~~~ent, the delivery may be ~eri"onned at temporally distinct times.
For instance, the asset may be delivered at different asset delivery spots on a cormion network channel.
In a further arTax~gernent, the asset may be delivered on two or ~~~~~
tlifferent network channels and/~r at two or more different times. ln this regard.s the asset may be delivered to a desired audience rather than at a specific time and/or ~~i a specific channel. Asset providers may intcTactively bid for sÃiclt, audience impressions wid/or provide predefiiied ad campaigns such that auctioning may be ~~rforfned in an automated process.
According to another aspect of the invention, a just-in-time auctioning method is provided for auctioning upcoming asset delivery spots in a broadcast content str~wn. '1V.he utility includes obtaining at least one audience characteristic for an upcoming asset delivery spot in a broadcast content stream and identifying ad campaigns that have targeting criteria corresponding with the at least. one audience characteristic. From the iclentified. ad ~~paigns, a wiauiing bid is identified and an asset associated with the winning ad campaign is delivered d-Luing the asset delivery spot. This process may be repeated for separate asset delivery spots. Altemativelyy, this process may be repeated for multiple parallel asset delivery spots for a common asset delivery spot.
Once an asset associated with the wim.Ling ad campaign is delivered, at least one parameter of tYiat winning ad campaign may be adjusted based on the delivery of the asset. Such a paraineter may in~l-uÃi~ the number for impressions delivered for ttae winning a(i campaign. In. another arrangement, adjusting ttie parairn~~~~ may include adjusting the costs (e.g., total costs) for the winning ad campaign, For instance, if an ad campaign has an available budget of one thousand dollars and the delivery of the asset r~~~ the asset delivery spot costs two liun~.re~. dollars, the two h~.~.~ e~.
dollars may be subtracted ~~om the available budget to produce a rernaining total of eight hundred dollars for future auctions.
In addition, information may be obtained regarding a number of network -users who receive the asset such that, for example, the cost for delivery the asset may be based at least in part c~ii the number of network users Wlio receive the asset. In such an arTar~getnent4 the cost of delivering the asset may be determined by multiplying the number of network users who receive the asset with a bid price associated with the winning ad campaign. Obtaining information associated. with r~~~~~ of network users who receive the asset may include obtaining statistical information and/or obtaining reports from UEDs of the network users who actually receive the asset.
The step of identifying ad ~~~paigns may further in~lu(ie identifying constraints within the campaigns and correlating those ~onstragnts, with the upcoming asset delivery spot. For instance, ad campaigns may have temporal targeting critcTia that correspond with time(s) the a:sset. providers wish to deliver their assets. Such temporal targeting criteria may include time of days, days of the week, start ~id ~~id time (e.g., 8:00 ain to 5:00 pm), mini~ium separation between repeated deliveries of the same asset etc. Further, constraints, may include ad campaigns, having progranu-ning targeting criteria that corresponds with ~,~rogranima~g associated with a given asset delivery spot.
Sucli prograniming targeting criteria may include ratings, inclusions and or exclusions or particulairly networks, program ratings (e.g., all viewers, adult content, etc.).
The process of determining a winning ad campaign may include multiplying a bid price associated with each ad ~ainpaign associated with the asset delivery spot to detennine an offering price for each ad campaign for the asset delivery spot.
In one aeT~geinent, the winning ad cainpaign may be selected based on the highest offering price. It will be further appreciated that the estimated audience may be separated into different estimated audiciices by demographic groups and a different bids of different ad campaigns may be multiplied by different estimated audiences to determine offering prices for each ad canipaigil. In this ~~~~rd., different ad campaigns may be biddiiig on different demographic groups.

BRIEF DESCRIPTION OF THE DRAW ~~ ~S
Fig. I illustrates major components of a cable television netw-ork.
Fig. 2 illustrates band-,xidth usage that is dynamically determined on a geographically dependent basis via networks.
Fig. 3 illustrates asset insertion as acconiplished at a headend.
Fig. 4 illustrates an exemplary audience shares of various net"vorlcs as may be used to set asset delivery prices for future breaks associated witb the program.
Fig. 5 illustrates delivery of assets to different users Watcbing the saine programmigig channel.
Fig. 6 illustrates audience aggregation across.
Fig. 7 illustrates a virtual channel in the context of audience aggregation.
Fig. 8 illustrates targeted asset insertion being implemented at Customer Premises Equipment (CJEDs).
Fig. 9 illustTates asset options being transmitted fTc~m a headend on separate asset channels.
Fig. 1.0 illustrates a messaging seÃ1uerz~~ ~~~~en a t?ED, a tietWork pl~tfonns ~~id a traffic and billing (T&B) sys~~in.
Fig. 11 is a flow c1iart illustrating a process for implementing timemslot and targeted impression buys.
Fig. 1.2 illustrates exemplary sequences associated with breaks on programming channels.
1~i& 13 illustrates an application that is supported by signals from 1.JEDs and which provides targeted assets to users of one or more channels within a network.
Fig. 14 illustrates the use of asset channels for providing assets slurigig a break of a programming channel.
1~ig, 15 illustrates a reporting system., ~

Fig. 16 illustrates an auctioning plalfonn incorporated into a targeted asset systÃ~x~.
Fig. 17 is a flow chart illustrating a first auction technique.
Flg, 18 is a flow chart illustrating a second auction technique.
Fig. 19 is a. flow chart illustrating a third a-Liction technique.
DETAILED DESCRIPTION
The present invention relates to various s~ctLire and functionality for delivery of targeted assets, classification of network users or consuming pattems, and network monitoring for use in a communications netWorko as well as associated business methods.
The invention is applicable with respect to networks where content is broadcast to network users; that is, the content is made available via the network to multiple users without being specifically addressed to individual user nodes in po1nt-to-point fashion. hi this regard, content may be broadcast in a va-raety of networks including, for example, cable and satellite television networks, satellite radio netw-orks, IP
networks used for multicasting content and networks used for podcasts or telephony brrsadcastslmultacastfi.
Content may also be broadcast over the airwaves though, as will be understood from the description below, certain aspects of the invention make use of bl-d1rect1onal communication channels -whicl~ are not readily available, for exai-nple9 in connection with conventional airwave based televisions or radios (a.e., sucb. communication would involve supplemental communication systems). In various contexts, the content may be coiisu~ed in real time or stored for subsequent consumption. Tltus7 while specific examples are provided below in the context of a cable television network for ~umoses of illustration, it will be appreciated that the lnvenklon is not limited to stich contexts but, rather, has application to a variety of network-s and transmission modes.
The targeted assets may include atiy type of asset that is desired to be targeted to network users. It is noted that such targeted assets are sometimes referred to as "addressable" assets (thou,,li, as will be understood from the description below, targeting ~aii be accompllsl~ed without addressing in a pointLLto4poant seiise). For exatnple, these targete~.~ assets may include advertisements, 1n~emal marketing (e.g,, lnfonnatlon about network promotions, scheduling or upcoming events), public service am~ou~~~ents, weather or emergency information, or programming. The targeted assets may be independent or included in ~~onlenx stream Witb other assets such as untargeted network progran-imin,g. In the latter case, the targeted assets may be interspersed with untargeted progranuning (eag., provided during ~~ogran1m1ng breaks) or may otherwise be combined with the programming as by being superimposed on a screen portion in the case of video proga~ming. In the ÃlescrilatiÃ~ii below, specific examples are provided in the context of targeted assets provided during breaks in television programming. VVhi1e this is an important com~ierc1a1 implementation of xlie 1n~ention.2 it will. be appreciated that the invention has broader application. 'l"hus, dlstlaictions below between "programming" and "assets" such as advertising sliould giot be Lmderstood as limiting the types of cor~~~iit that may be targeted or the contexts in which such content may be provided.
The following description is divided into a number of secxioris. En the 14 Introduction sectlÃ~ilp the broadcast network antl network programming environments are first described. Thereafter, an ow~t-vieW of the targeted asset environment is provided 1nÃ;luÃl1ng a Ãl.iscusslÃ3n. of ~efta1n shortcomings of t1ie conventional asset delivery paraÃligrii. The succeeding section describes components of a targeted asset system lrt, accordance with aspects of the present invention highlighting advantages of certain preferred lmplerner~tatlons thereof Finally, the last section describes various structure and fmictionality for iflnplementing auctioning of delivery spots and/or corau-nerclal impressions.

:ItQ INTRODITC Y"ION
A. Broadcast NetWorks "1"he present in~ention. has particular application in the context of networks primarily used to provide broadcast content, herein termed broadcast networkso Such broadcast networks generally involve synchronized distribution of broadcast content to multiple users. However, it will be appreciated that certain broadcast networlts are not limited to synchronously pushing content to multiple users but can also be used to deliver coaitent to specific users, including on a user pulled basis. As noted above, examples of broadcast networks include cable televisi~ii networks, satellite television lietW orks, and satellite radio networks. In addition, audio, video or other content may be broadcast across Internet protocol ~id telepbony networks. In any s-ucli networks, it may be desired to insert targeted assets such as advertisements into a broadcast stream.
Exaniples of broadcast network-s used to delivery content to specific users include broadcast networks used to deliver on demand content such as VOD and podcasts. The present invention provides a variety of fimctionality in this regard, as will be discussed in.
detail below.
For purposes of illust-ratlon, the invention is described in some instances below in the context of a cable television network implementation. Some major components of a cable television network 100 are depicted in Fig. 1. In the illustrated network 100, a headend 104 obtains broadcast content from any of a number of sources 101-1 Q3o Additionally, broadcast content may be obtained from storage media 1 05 such as a via a video serve.r. 17he illustrated sources include an antenna 101, for example, for receiving content via the ainvaves, a satellite disli 1.02 for receiving content via satellite ce~~imunicafionsv and a fiber link 1 03 for receiving content directly from studios or other content sources. It will be appreciated tb.at the illustrated sources 101 m 1~3 and 105 are provided for purposes of illustTatia~ii and other sources may be utilized.
`1'lie la~ad~~id 104 processes the received content for transmission to network users.
Among other things, the headend 104 may be operative to amplify, convert and otherwise process the broadcast ca~iitent signals as well as to combine the signals into a common cable for transmission to netNv~rk users 1 07 (although graphically depicted as households, as described below, the system of the present invention can be used in implementations where individual users in a housebold are targeted). It also is not necessary t.bat, the target audience be composed households or houseliold ~nembers in any sense. For example, the present invention can be used to create on-th~-fly customized presentations to students in distributed classrooms, e.g., thus providing examples which are more relevant to each student or group of students Witliin a presentation being broadcast to a wide range of students. The headend also processes signals from users in a variety of contexts as described beloW, The headend 1 04 may thus be thought of as the cÃ~iitrol center or local control center of the cable television network 100.
Typically, there is not a direct fiber li~ik from the headend 1 04 to the customer prea-ni~~~ equipment (CPE) 108. Rather, this connection generally involves a system of feeder cables and drop cables that def-irae a number of syst~ subsections or branches.
This distribution netdvork may include a number of nodes 109. '1'he si~ial may be processed at these iiodes 1 09 to insert localized content, filter the locally available channels or otherwise control the content delivered to users in the node area.
The resulting content within a node area is tyTically distributed by optical and/or coaxial links 106 to the premises of particular users 107. Finally, the broadcast signal is processed by the UED 108 which may include a television, data temiinal, a digital set top box, DVR or other terminal esluipr.~ent, It will be appreciated that digital or analog signals may be involved in this regard.
Users ernploy the network, and network operators derive revenue, based on delivery of desirable content or programming. The stakeholders in. this regard include prograr.uning providers, asset providers such as advertisers (who may be the same as or different than the programniing providers), network operators such as Multiple Syst~ins Operators (MSOs), and userS----or viewers in the case of television netWorks.
Progranuning providers ir#cl~ide, for example: networks who provide series and oth.er pr~grammitig, including on a national or international basis; local affiliates who often provide local or regional programming; studios who create and market content including movies, documentaries a~id the like; ~id a variety of otl~er content owners or providers.
Asset providers include a wide variety of manufactu.rers, retailers, service providers and public interest gToups interested in, and generally willing to pay for, the ~~po-rt~iiity to deliver messages to users on a local, regional, national or international level. As discussed below, such. assets iticlude: conventional advertisements; tag content such as ad tags (Whieh may include static graphic overlays, animated graphics files or even realntime video and audio) associated with tl~~ advertisements or other content; banners or other content superimposed on or otherwise overlapping programminga product placement; and other advertising mechanisms. In addition, the networks may use insertion spots for lntemal marketing as discussed above, and the spots may be used for public service announcements or other iionRadvertising contenta Network operators are generally responsible for delivering content to users and otherwise operating the nehvÃsrks as well as for contracting with the networks and asset providers and billing. Users are the end.
cons~~~rs of the content. Users may ernplÃsy a variety of types of UEDs including television, set top boxes, lP:~DTm devices, data terminals, satellite delivered video or auÃ11~~ to an automobile, appliances (such as refrigerators) with bullt-in televisions, efic.
As described below, all of these stakeholders have an interest in improved delivery of content including targeted asset delivery. For ex~iple, users can thereby be exposed to assets that are more likely of interest and can cÃ~~~i-nue to have the costs of programming sulssad.lzed or wholly bom~ by asset providers. Asset providers can benefit from more effective asset dellvery and greater return on their investinent.
Network operators and asset pro'%,riders can benefit from increased value of the network as an asset delivery mechanism and, thus, potentially enhanced revenues. The present invention addresses all of these interests.
It will be noted that it is sometimes unclear that the interests of all of these stakeholders are aligned. For example, it may not be obvious to all users that they bene-fit by consuming such assets. Indeed, some users may be willing to avoid consuming such assets even with ati understanding of the assoclated costs. Nehvorlà operators and asset providers may also disagree as to how programming should best be disffibuted., how asset delivery may be associated with the p~~~~mmln.g, and how revenues sl~ould be shared.
As described below, the present invention provides a mechanism for accommod.atiaig potentially conflicting interests or for enhancing overall value such that the interests of all stakeholders can be advanced.
Assets can be provided via a variety of distribution modes including realRtlme broadcast distribution, forwarÃl-aeid-stÃsrey and Ãsnsderf.and delivery sudh as VOD. Reiil-time broadcast delivery involves syriÃ:lrÃanous delivery of assets to multiple users sucli as the coaiveiitaonal paradigm for broadcast radio or television (e.g., airwave, cable or satellite). The forW ard.mand.mstore mode involves delivery of assets ahead of time to UEDs with substantial storage resources, e.g., a DVR or data terminal. The asset is skored for later d.isplay, for example, as prompted by the user or controlled according to logic resident at the UED widEor elsewhere in the commui1icatioits network. The on-demand mode involves and.iviÃluallzesl delivery of assets lTÃsm the network to a user, often on a payyper-vieW basis. "t"he pres~iiI in~ention. can be utilized in connection with any of these distribution modes or others. In this regard, important features of the present invention can be implemented using conventional UEDs wifil~out requiring substantial storage resources to enhance even ~ealwtime broadcast progranuningY for analog and digital users.
The amount of programming that can be delivered to users is limited by the available prograniniing space. 'I`hifi3 in tum, is a function of bandwidth.
Thus, for example, cable television networks, satellite television networks, satellite radio networks, and other networks have certain bandwidth limitatioiis. In certain broadcast networks, the -ivailable bandwidth niav be divided into bandwidth portions that are used to transmit the programming for individual channels or stations. In addition, a portion of the available bandwidth may be utilized for biTdirectional messaging, metadata transmissions and other network overbead. Alternately, such bi-direcIional com-rnu.ricatgon may be accommodated by any appropriate communications channels, including the use of one or more separate communications networks. The noted bandwidth portions may be defined by dedicated segments, e.g., defined by frequency ranges, or may be dynamically configured, for example, in the case of packetized data networks. As will be described below, in one impI~.entation, the present invention uses available (dedicated or oppoatLu-iÃsticaIly available) bandwidth for substantially real time transmission of assets, e.g., for targeted asset deliwc-ry with respect to a defined asset delivery spot. In this impl~enxation, bi-directional communications may be accommodated by dedicated messaging baaidwidth and by encoding messages within bandwidth used for asset delivery. A DOCSIS pa~~ or certain TELCO solutions using sWitched. IP may be utilized.
for bimdirectional comnitinications between the headend and I~ EDs aiid asset delivery to the t?EDs, including res.l-time asset delivery, in the systerns described below.

B. Scheduling What programming is available on particular channels or other bandividth segments at particular times is determined by scheduling. I`1ius, in the context of a broadcast television network, individual pr~grarmning networks, associated with particular progra:anming c1iannels, will generally develop a programming schedule well into the future, e.g., week-s or months in advance. This la~~graniming schedule is generally published to users so that users can find programs of interest. In addition, this prograrnming schedule is used by asset providers to select desired asset delivery spots.
Asset delivery is also scheduled. That is, breaks are typically built into or otherwise provided in programming content. In the case of recorded content, the breaks are preyd.e~aned. Eveii in the case of live broadcasts, breaks are built-in.
Thus, the number and duration of breaks is typically known in advance, though the exact timing of the breaks may ~ ai-y to some extent. However, this is not always ttie case.
For example, if spor&;irig events go into overtime, the number, duration and timing oi"~~eak-s may vary dynaniicaliy. As discussed below, the system of the present invention can handle ~eaiy i0 time delivery of assets for updated breaks. In connection with ~egLiiariy sched-Liieri breaks, as discussed below, defined avail windows estabiisb the time period during which certain breaks or spots occur, and a cue tone or cue message signals the beginning of such breaks or spots. In practice, an avail window may be as long as or longer thati a ~~~graTn and include all associated breaks. [rt,rleed, avail wiaidows may be several hours long, for ~xarnple, in cases where audience dernÃ~~apMcs are not expected to change significantly over large programming blocks. In this regard, an MSO may merge multiple avail windows provided by pr~~Tammia~g networks.
More specifically, a break- may include a series of asset delivery spots and the content of a break may be deteranined by a number of entities. For example, ss~~~e asset delivery is distributed on a basis coextensive with ~etw-ork programming, e.g., on a national basis. This asset delivery is conventionally scrieduied based on a timed playlist.
"1'hat is, the insertion of content is centrally coaitrolleÃi to insert assets at defined times.
Accordingly, the programming and national asset delivery may be provided by the progranunflng networks as a continuous content st~eain without cues for asset insertion.
For example, prime-time prcagramiyiing on the ~najor networks is often principally provided in this fasbiono In other cases, individual spots within a break are allocated -for Regional Operations Center (ROC), affiliate, super headend or local (head~nd, zone) content. in these cases, a cue tone or message identifies the start of the asset delivery spot or spots (a senes of assets in a break may all trigger from one cue). The cue generally occurs a few seconds before the start of the asset delivery insca-tion opportunity and may Ã~~c-ury for example, during programn-iiiig or dufing the break (e.g., during a national ad). Tlle system of the present invention ~ati be impieanented at any or all levels of this hierarchy to allow for targeting with respect to national, regional and local assets. In the case of regional or local targeted asset delivery, synchronous asset options (as discussed below) may be inserted iiito designated bandwidth in response to cues. In the case of national asset delivery, network signaling may be extended to provide signals identifying the start of a national spot or spots, so as to enable the inventive systern to insert syncIroiioLas national asset options into designated bandwidth. For ~xainple, such s.ignali-nl; may be encrypted for use only by the invetiti~~ targeted asset systern.
Network operators or local network affiliates caii generally schedule the non-national assets to be iiicluded within defined breaks or spots I'or each ad-supported channel. C'ort,ventioiialiyy, this scheduling is finalized ahead of time, typically on a daily or longer basis. The scheduled assets for a given break are then typically inserted at the headend in -response to the cue tone or message in the programming stream.
Thus, for example, where a given avail window includes three breaks (each of Whicb. ~nay include a series of spots), the scheduled asset for the first break is inserted in response to the first cue, the scl~~d-LileÃI asset for the second break is inserted in response to the second cue, and the scheduled asset for the third break is inserted in response to the third cue. i:f a cue is missed, all subsequent assets within an avail window may be thrown off It will be appreciated that such static, daily scheduling can be problematic.
For example, the progranu-ning schedule can often change due to '~realcir~g, news, ripple effects ftom schedule over-runs earlier in the day or the nature of the programming.
For ~~atupley certain live events such as sporting events are difficult to precisely selieÃlule. In such cases, static asset delivery schedules can result in a mismatch of scliesluled asset to the associated programming. For example, when a high value prograniming event such as a certain sporting event runs over the expected prograin length, it may sometimes occur tliat assets intended I'c~~ another program or valued for a smaller audience may be shown when a higher value or better-tailored asset could have been used if a more dynamic scheduling regime were available. 17he asset targeting systert, allows for such dynamic sclieduling as will be discussed in more detail below. The asset targeting system ~~i also accommodate evolving standards in the field of dynamic scheduling.
C. The Conventional Asset DelivM Paradigm Conventional broadcast networks may inclLide asset-supported and Isrernium content charanels/n~tworks. As noted above, programming content generally comes at a substantial cost. That is, the progranuning providers expect to be ~ompensated. for the programming that they provide wlich has generally been developed or acquired at signi-ficank cost. That compensation may be generated by asset delivery revenues, by. fees paid by users for premium cbannels, or some combination of the two. In some cases, funding may come from another source sucb as public funding.
In the case of asset-supported networks, the conventional parad.igni involves timey slot buys. Specifically, asset providers ~~~~rallv id,entifv a particular p~~~am or tim.e-sl~~ on a particular network where they desire their assets to be aired. The cost for the airing of the asset depends on a itumber of factors, but one primary factor is the size of the audience for the programming in connection with which the asset is aired.
"I"lxus, the standard pricing model is based on the cost per thousand vieWers (CPM), though other factors stich as demographics or audience composition are involved as discussed below.
The size of the audience is generally determined based on ratings. The most common benchmark for establishing these ratings is the system of Nielsen Media Research Corporation (Nielsen). One technique used by Nielsen involves monitoring the viewing habits of a presumably statistically relevant sampling of the universe of users. Based on an analysis of the sample group, the Nielsen system can estimate what portion of the audience particular programs received and, from this, an estimated audience size for the program can be projected. 'Ihus, the historical performance of the particular program, for example, as estimated by the Nielsen systerns may be used. to set asset delivery prices for future breaks associated witli that program.
In practice, this results in a small r~uTraber of prograniming networks being responsible for generating a large portion of the overall asset revenues. This is graphically depicted in Fig. 4 which generally illustrates this phenomenon, although it is not based on actual nuinbers. As shown in Fig. 4, it is often the case that three or four progTamyning networks out of many available p~ogranuni~g networks gamer very large sbares whereas the ~~naining prog am-min~ networks have small or negligible share.
Indeed, in some cases, many programming networks will have a share that is so small that it is difficult to statistically characterize based on typical Nielsen sampling group sizes.
In these cases, substantial asset revenues may be generated in connection with the small number of pra~animiiig networks having a significant share while very little revenue is generated with respect to the other programming networks. This is true even though the other prograrnming networks, in the aggregate, may have a significant nuanber oI'users in absolute tennse '1'h-us, the conventional paradigm often fails to generate ~~~~llues ~~~ensurate with the size of the total viewing audience serviced by the network operator. As discussed below, this is a missed revenue opportLuiÃty that can be addressed in accordance witb the present invention.
1; As noted above, the pricing for asset delivery depends oti the size of the vieWi~~g audience and certaiii other factors. One of those factors relates to the demographics of iiiterest to the asset provider. In this regard, a given program will generally have a nuniber of different ratings for different demographic ~atego fies. T hat is, the program generally has not only a household rating, which is measured against the universe of all households with televisions, but also a rating for different demographic categories (e.g., males 1 8-24)g measured against the universe of all members of the category who have televisions. '~'hus, the program may have a rating of 1 (l~`~) overall and a rating of 2 (2%) for a particular category. Typically, when asset providers buy a time-slot, pricing is based on a rating or ratings for the categories of interest to the asset provider. 17his results in significant inefficiencies due to poor matching of the audience to the desired demographics.
Conventionally, asset insertion is accomplished at the headend. rMs is illustrated in Fig. 3. ln. the illustrated system 300, the headend 302 includes a program feed 304 and an asset source 306. As noted above, the program feed 304 may be associated with a variety of programming sources such as video storage, an antenna, satellite dish or fib~.~
feed ftom a studio or the like. "1`he asset source 306 may include a tape library or other storage system for storing pre-recorded assets. A pl~.tf~snra associated with the headend 302 -- in this case, denoted a selector 308 -- inserts progr~.mi~g frorn the p~ograrn~~ed 304 and assets i'xorn the asset source 306 into the video st~eairn of an individual c1~annel 3 10. l'bis is done for each channel to define the overall content 312 that is distributed to subscribers (or at least to a node filter). l`ypically, although not necessarily, the selector 308 effectively toggles between the program feed 304 and the asset source 306 such that the programming and assets are inserted in altemating, non-time overlapping fashion.
Thus, as shown in Fig. 3, a particular channel. may include a time segment 314 of programming followed by a cue tone 316 (which may occur, for example, during a programming segment, or during a time period of an asset provided with the programming stream, just prior to an insertion opportunity) to identify the initiation of a break 318. In response to the tone, the selector 308 is operative to insert assets into the prog arnming stream for that channel. At the conclusion of the break 318, the selector 308 ~etums to the program feed to infieTt a further programming segment 320.
An example of a timeline in this regard is shown in Fi& 15, This content 312 or a filtered portion thereof is delivered to LJEDs 322. In the illustrated embodiment the UED 322 is depicted as including a signal processing component 324 and a television display 326. It will be appr~~~-iated that these coan~oneiits 324 and 326 may be embodied in a single device and the nature of the functionality may vary. In the case of a digital cable tiser, the sigaial processing compofl-ient 324 may be incorporated into a digital set top box (DSTB) 1"~r decoding digital signals.
Such boxes aTe typically capable of bi-directional messaging with the headend 302 which uyill be a significant consideration in relation to fimcliogiality described below.

11a SYSTEM ~~ERNrTEW
A. The 'C'argexe-d AsseÃ. Del~~~ En~~~omnent Against this backdrop described in the context of the conventional asset delivery p~ad~gin, a system embodying the present invention is described below. The inventive system, in the eanl~odiments described belssWo allows for delivery of targeted assets such as advertising so as to address certain shortcomings or inefficiencies of conventional broadcast networks. Generally, ~uch targeting entails delivering assets to desired groups of individuals or individuals having desired characteristics. These characteristics or audience classification parameters may be defined based on personal information, demographic inl'Ã~-rmation, psychographic in1"ormatgon, geographic in1"ormation, or any other inl"anna~~on that may be relevaait to an asset provider in ~~entifyin~ a target audience. Preferably, such targeting is pr~gra~~ ~~id~en~~~t in recognition that progranuning is a higl-dy imperfect mechanism for targeting of assets. For ~xamplee even if user ~ialysas ~iid~cates that a particular program has an audience coinprised sixty ~~~~iit of women, and women comprise the target audience for a particular asset, airing on that program will result in a forl~r percent mismatch. That is, forty percent o1"the users potentially reached may not be of interest to the asset provider and pricing niay be based only on sixty percent of the total audience. Moreover, ideally, targeted asset delivery would allow for targeting with a range of granularities including very fine granularities.
For example, it may be desired to target a g~oup, sucli as based on a geographical grouping, a household c1~ar~~terazataon or even an individual user e1~aracterization. 'T'he present invention accommodates program independent targeting, targeting with a high degree of granularity and targetiiig based on a variety of different audience classifications.
Figs. 5 and 6 illustrate two different contexts of targeted asset delivery supported in accordance with the asset targeting system. Speci~cally, Fig. 5 illustrates the delivery of different assets, in this case ads, to different users watching the same programming chamel, which may be refeaTed. to as spot optimization. As sb.owti, tliree di'i'e-rent users 500-502 are depicted as watching the same programming, in this case, denoted "Movie of the Week." At a given break 504 the users 500-502 each receive a d.itlerent asset package. Specifically, user 500 receives a digital music player ad and a movie promo, user 501 receives a luxury car ad and a health insurance ad, and user 502 receives a minivan ad mid a department store ad. Aitemately, a single asset provider (e.g.4 a motor vehicle company) may purchase a spot and then provide different asset aptioiis for the spot (e.ge, sports car, minivans, pick-up trucks, etc.). Similarly, separate advertisers may collectively purchase a spot and then provide ads for their respective products (e.g., where the tar,~~t audiences of the advertisers are complementary). It will be appreciated that these different asset packages may be targeted to different audience demographics. In this manner, assets are better tailored to particular viewers of a given program who may fall into different demographic groups. Thus, spot optimization refers to the delivery of d.i5erent assets (by one or multiple asset providers) in a given spot.
Fig. 6 illustrates a different context of the present invention, which may be termed aud.i~~~~e aggregation. ~~i this case, three different users 600Y602 viewing different progranis associated with different channels may receive the s~.e w"et or asset pack-age.
In this case, each of the users 6Ã30-602 receives a package including a digital music player ad and a movie promo in connection wgth. breaks s.ssociated witb their respective ch~~iiels. 'rhough the users 600~602 are s.hoWn as receiving the s~~e asset package for purposes of illustration, it is likely that different users will receive different combinations of assets due to differences in classilicatioii paraineters. In this n$~tiner., users over multiple channels (some or all users of each channel) can be aggregated (relative to a given asset and time window) to define a virtual channel having significant user numbers matching a targeted audience classification. Amoiig other things, such audience aggregation allows for the possibility of aggregating users over a number of low share dbannels to define a significant asset delivery opportunity, perhaps on the order of that associated with one of the high share networks. This can be accomplished, in accordance with the present invention, using equipment already at a user's premises (i.e., an existing I1ED). Such. a virtual channel is graphically illustrated in Fig. 79 though this illustration is not based on actual numbers. Thus, audience aggregation refers to the delivery of the same asset in different spots to define an aggregated audience. 1."hese different spots may occur within a time window corresponding to overlapping (conflicting) programs on different channels. In this manner, it is likely that these spots, even if at different times witliin the wifl}doW, will not be received by the same users.
Such targeting iiicluding batli spot optimization and audience aggregation can be implemented using a. variety of architectures in accordance with the asset targeting system. Thus, for exaniple, as illustrated in Fig. 8, targeted asset insertion can be iraiplemented at the UEDs. This may involve a f~~ardsand6store fuiictionality.
As illustrated in Fig. 8, the UED 800 receives a progranm-iing stream 802 and an asset delivery stream 804 from the headend 808. These streams 802 and 804 may be provided via a comma~ii signal link such as a coaxial cable or via separate communications 1inlcs.
For example, the asset delivery stream 804 may be ~~ansmi~~ed to the UED 800 via a designated segrnent, e.g., a dedicated frequency range, ai ~e available 1~andWiÃith or ~Yia a programming channel that is opportunistically available for asset delive-rv, e.g., when it is otherwise off air. The asset delivery stream 804 may be provided on a continuous or intermittent basis and may be provided cr~ncurTeiitly with the ~rogranuning s~earn 802.
In the illustrated exaniple, the programming stTeam 802 is processed by a p~~~~Tn decoding unit, such as DSTB, and progranmiing is displayed on television set 814.
Al~emativelyy, tthe programming s~eam. 802 may be stored in prograniming strsrage 815 for UED insertion.
In the illaastra~ed implementation, the asset, together with metadata identifving, for ~xatnp1e, any audience classification parameters of the targeted audience, is stored in a designated storage space 806 of the ITED 800. It will be appreciated that substantial storage at the UED 800 may be required in this regard. For ~~aniple, such storage may be available in connection with certain digital video ~ecordcr (DVR) -unitso A
selector 810 is implemented as a processor running logic on the UED 800. The selector 81.0 functions analogously to the headend selector described above to identify breaks 816 and insert appropriate assets. In this case, the assets may be selected based on classification parameters of the household or, more preferably, a user within the household.
Such information may be stored at the UED 800 or may be det.e-rmined based on an analysis of viewing habits such as a click str~eain from a remote control as will be described in more detail beloW. Certain aspects of the present inventiorf, can be implemented in such a UED
insertion envg~onmento In Fig. 9, a different architect-ure is employed. Specifically, in Fig. 9, asset options transmitted from headend 910 synchronously with a given break on a given channel for which targeted asset options are supported. The 1_lED 900 includes a channel selector 902 whicli is operative to switch to an asset channel associated with a desired asset at the begin-ning of a break ~id to return to the programming c1iaiinel at tlie end of the break. 'I'he channel selector 902 may hop betweeai ~~annelS (between asset channels or bet~~~~i an asset channel and the progranmlin,~ ~hamiel) d-uring a break to select the most appropriate assets. ln. this regard, logic resident on the UED 900 controls ~ueb.
haappiiig to avoid switching to a channel Nv1~ere an asset is al~~ady in progress. As described below, this logic can be readily implemented, as the schedule of assets on each asset channel is krfloWii. Preferably, all of this is implemented invisibly from the perspective of the user of set 904. The different options may be provided, at least in part, in. connection with asset channels 906 or other bandwidth segments (separate ftom programming channels 908) dedicated for use in pro-visli~~ such options. In addition, certain asset options may be i~~~~ed into the current programming channel 908.
Associated functionality is described in detail below. The architecture of Fig. 9 has the advantage of not requiring substantial storage reso-urces at the UED 900 such that it can be immediately iinpl~~nted on a wide scale basis using equipment that is already in the fieldo As a ftirther alt~mativeY the detenninatia~~ of which asset to show may be made at the headend. For example, an asset may be selected based on voting as described below, and inserted at the headend iflito the progTanmlin,~ channel without options on other asset.
channels. '1Tiis would achieve a degree of targeting but without spot optimization opportunities as described above. Still further, options may be provided on other asset channels, but the selection as between those channels may be deterinined by the headend.
For exanple, infr~imation about a household or user (e.g., brand of car Ã~wned., magazines subscribed to, etc.) stored on. the headend may be used to match an asset to a household or user. That inf~rmatiofli, which may be termed "marketing labels," may be used by the headend to control Which asset is selected by the UED. For example, the UED
may be instructed that it is associated with an "ACME lsreferred.'F customero NN'1~en an asset is disseminated with ACME preferred metadata, the UED may be caused to select that asset, tliereby overriding (or significantly factoring with) any other audience classification considerations. However, it will be appreciated that such operation may entail certain ~~~~cems rel.atlrig to sensitive information or may compromise audience classification based targeting in other respects.
A significant oplsoTtunlty thus exists to better target users whom asset providers may be willing to pay to reach and to better reacb, b,ard-tÃ~~~each users.
However, a number of challenges remain with respect to achieving these ob~jectives including: how to provide asset options within network bandwidth limitations and without requiring substantial storage requirements aiitl new equipment at the user's ~reinlses;
how to obtain sufficient information for effective targeting while addressing privacy cr~~~ems; how to address a variety of business related issues, such as pricing of asset delivery, resulting 1'xom availability of asset options and atten(lant contingent delivery; and how to operate effectively within the context of existing network structure and systems (e.g., across node filters, using existing traffic and billing syst~ins, etc.).
From the foregoing it will be appreciated that various aspects of the invention are applicable in the context of a vanety of networks, including broadcast networl~.s. In the following discussion, specific implementations of a targeted asset system are discussed in the context of a cable television network. Though the system enhances viewing for both aiialog and digital users, certain functionality is ca~iivenlent:ly implemented using existing DSTBs. It will be appreciated that, wblle these represent particularly advantageous and ~onirne-rcially valtiable impleanentations7 the invention is not limited to these specific i-inplementations or network contexts.
B. 5~y:~tra Arcliitecture ln one implementation, the systein of the present invention involves the t~wismission of asset options in time aligmnent or synchronization with other assets on a programming channel, where the asset options are at least partially provided via separate bandwidth segments, e.g. channels at least temporarily dedicated to targeted asset delivery. Although such options may typically be transmitted in alignment with a break in programming, it may be desired to provide options opposite continuing ~~~~anuning (e.g., so that only subscribers in a specified ~~~grapbic area get a ~eatli~~
announcement, an emergency announcement, election results or other local information while others get uninterrupted programming). Selection as betn%een. the available options may implemented at the user's premises, as by a DSTB in this irxiplernentation. la this manner, asset options are made available for better targeting, without the requirement for substantial storage resources or eq-Liilsm~~t upgrades at the user's premises (e.g.x as might be reqluirecl for a frsrWard-and-store archi~~~ture). Indeed, existing DSTBs can be configured to execute logic for implementing the system described below by dowailoadang and/or preloading appropriate logic.
Because asset options are synchronously transmitted in this implementation, it is desirable to be efficierat in iel.entil'~.r~~ available banclwidlb and in using that bandwidth.
Various fwtctissraality for improved '~aiidwidth identification, e.g.y ldetitifying lsa.rs.Ã1widtli that is opportunistically available in relation to anode filter. Efficient use of available bandwidth involves both optimizing the duty cycle or asset density of ~~vaf1able bandwidth segment (i.e., how much time, of the time a bandwidth s~gi-nent is available for use in transmitting asset options, is the segment actually used for transmitting options) and the value of the options transgnittel. The forrner factor is addressed, among other tl-iingss by improved scheduling of targeted asset delivery on the asset channels in relation to scheduled breaks of the programming ch~iiels.
The latter factor is addressed in part by populating the available bandwidth spots 2Ã3 with assets that are most desired based on current network conditions.
These most desired assets caii be determined in a variety of ways including based on c~.~nventissnal.
ratings. ln the specific implerflicfl~tatiort, described below, the most desired assets, are determined via a process herein ter~~ed voting. Fig. 10 illustrates an associated messaging sequence 1 040 in this regard as between a UED 1 002 such as a DS'TB, a network platform for asset insertion such as a headend 1004 and a traffic and billing (T&B) systeni 1006 used in the illustrated ~xamp1e for obtaining asset delivery orders or contracts and billing for asset delivery. It will be appreciated that the functionality of the T&B system 1006 may be split between multiple systems running on multiple pl~~forrns and the T&B system 1006 may be operated by the network operator or may be separately operated.
'rbe illustrated sequence begins by loading contract information 1008 from the T&B systern 1006 onto the headend 1.004. An interface associated with system allows asset providers to execute contracts for dissemination of assets based on traditional Iime-sloI buys (for a given progà am. or given time on a given network) or based on a certain audience classification in~~rination (e.g., desired dernogTalshics, psyc1~~graphicS, geography, and/or audience size). In the latter case, the asset provider or network may identify audience classification inI'onnation associated with a target audience, ne system 1006 uses this information to compile the contract information 1 008 which ideaitifies the asset that is to be delivered together with delivery pa~~ineters re,~~rdirzg when and to whom the asset is to be delivered.
The illustrated headend 1004 uses the contract information together with a scl~edule of breaks for individual networ~.s to compile an asset option list 1010 on a charnel-byachanneI wid breakyby-Is:reak basis. That is, the list 1010 lists the universe of asset options that are available for voting ~urposes for a given break on a given programming channel together with associated metadata i~~ntifving the target audience for the asset, e.g., based on. audience classification i~~~rmation. The transmitted list 10113 may encompass all su~~orted programming channels and may be transmitted to all participating users, or the list may be limited to a~~~~ or a subset of the supported ch~n nels e.g., based Ã~ii an input indicating the current channel or the most likely or frequent channels used by a particular user or group of users. The list 10 10 is transmitted from the headend 1004 to the UED 1002 in advance of a break for which options are listed.
Based on the list 1010, the UED 1002 submits a vote 1012 back to the headend 1004. More specifically, the UED 1002 first identifies the classification parameters for the current user(s) and perhaps the current channel being watched, identifies the assets that are available for an tipcoming break (for the current channel or multiple channels) as well as the target audience for those assets aiid determines a "fit" of one or more of those asset options to the c-Li.r~ent classificationo In one implementation, each of the assets is attributed a fit score for the user(s), e.g., based on a comparison of the audience classification parameters of the asset to the putative audience classi~cation.
parameters of the cuaTent user(s). This may involve how well art, individual user classification parameter matches a corresponding target audience parameter and.'or how many of the target audience parameters are matched by the user's classification ~~~~terso Based on these fit scores, the UED 102 issues the vote 1012 indicating the most appropriate asset(s). Any s-uitable information can be used to provide this indication.
For example, all scores for all available asset options (for the curretit charmel or multiple channels) may be included in the vote 1012. Al~ematively, tl-ie vote 1012 may identify a subset of one or more options selected or deselected by the UED 1002, with or without scoring information indicating a degree of the match and may further include channel infrsrmataon. In one iinplemea~tat1on, the 1~ead~nd 1Ã304 instructs I..;EDs (1002) to return fit scores for the top N asset options for a given spot, where N is dynamically configurable based on any relevant factor such as network traffic levels ~id size of the audience. Preferablv, this votiniz occurs shortly before the break at issue such that the voting more accurately reflects the current status of network users. In one implementation, votes are only submitted ft)r ttie p~ogramniatig channel to wh1cb.1be L;ED
is set, and votes are submitted periodically, e.g., every fifteen minutes.
The headend 1004 compiles votes 1012 ~om UEDs 1002 to determine a set of selected asset options 1014 for a given break on a supported pr~ogra.r~un1flig cb,aiinel. As will be understood from the rlescripÃ.1on. below, such votes 10 12 may be obtained from all relevant and participating UEDs 1002 (who may be representative of a, larger audience including analog or otherwise non-partic1pating users) or a statistical sampliiig thereof In addition, the headend 1004 ~~~emiines the amount of bandwidth, e.g., the number of dedicated asset option channels, that are available for transmission of options in support o1's. given break for a given programniiiig 6banneI.
Based on all of this 1nforination, the headend 1 004 assembles a flotilla of assets, e.g., the asset options having the highest vote values or the higliest weighted vote values where sucb. weighting takes into account value per user or other information beyond classification fit. Such a flotilla may include asset options inserted on the cu~~~it progTamniiiig c1~aimel as well as on asset channels, though different insertion processes and components may be involved for p~ogamm1ng cb.arm.eI and asset channel insertion.
It will be appreciated that some assets may be asseiiabled independently or largely independently of voting, for example, certain public service spots or where a certain provider has paid a premium for guaranteed delivery. Also,1n spot optimization co~.texts where a single asset provider buys a. spot and then provides multiple asset options for that spot, voting mav be unnecessary (though voting may still be used to select t1le options).
In one implernentat1on, the flotilla is assenibled into sets of asset options for each dedicated asset channel, where the time length of each set matches the length of the break, sucb that channel b.opplaig within a break is unnecessary.
Alternatively, the UED
1002 may navigate between the asset channels to access desired assets within a break (provided that asset starts on the relevant asset channels are fiviicb.roni~ed). However, it will be appreciated that the flotilla matrix (where columns include options for a given spot and rows correspond to channels) need not be rectangular. Stated differently, some channels may be used to provide asset options for only a portion of the break, i.e., may be used at the start of the break for one or more spots but are not available for the entire break, or a~ia~ only be used after one or more spots of a break have aired. A
list of the selected assets 1014 and the associated asset channels is then transmitted together with metadata identifying the target audience in the ill.ustTated. impl~entation, It will be appreciated that it may be unnecessary to include the inetadata at this step if the UED
1002 has retained the asset option list 1010. 'Fhlfi list 1014 is preferably transmitted shortly in advance of transmission of the asset 1016 (which includes sets of asset options for each rledlicated contact options channel used to support, at least in part, the break at, issue).
'Fhe UED 1002 receives the list of selected asset options 1014 and associated metadata and selects Whlch of the available options to deliver to the user(s).
For ~xample, this may iflivalve a comparison of the current audience classification paraineter values (which may or may not be the same as those used for purposes of voting) to the metadata associated with each of the asset options. 'l"he selected asset option is used to selectively switch the UED 1002 to the corresponding dedicated asset options channel to display the selected asset 10 16 at the beginning of the break at issue. One of the asset option sets, for example, the one comprised of the asset receiving the Mghest vote values, may be inserted into the p~~granuning chamiel so that switching is not required for many users.
Asfiuralng that the voting UEDs are at least somewhat representative of the universe of all -users, a significant clegr~~ of targeting is thereby achieved even for analog or otherwise ~~on-~partacgpatl~g users. In this regard, the voters serve as proxies for non-voting users.
'I'he U'ED 1 002 retums to the pra~~mimang cba.nnel at the conclusion of the break.
Preferably, all of this is transparent from the perspective of the user(s), i.e., preferably no user input is required. The system may be designed so that any user input overrides the targeting syst~in. For example, if tl-ie user chatiges channels during ab$ea.1g the change will be implemented as if the targeting s}ysteTn was not in effect (e.g., a ~om~nand to advance to the next channel will set the UED to the channel immediately above the current programming channel, without regard to any optaoiis currently available for that channel, regardless of the dedicated asset channel that is currently sourcing the television output).

In this system architecture, as in forWard-ansl-s~~~e architectures or any other option where selections between asset options are implei-nented at the UED, there will be some uncertainty as to how many users or households received any particular asset option in the absence of reporting. This may be tolerable from a business perspective. ln the absence of reporting, the audience size may be estimated based on voting data, conventional ratings analysis and other toolse Indeed, in. the conventional asset delivery paradzgni, asset providers accept Nielsen rating estimates and demographic inl`o-rmation together with market analysis to gauge return on iiidres~nent. However, this uncertainty is less than optimal in any asset delivery enviromnent and may be particularly problL-matic in the context of audience aggregation across multiple pra~graraimin~
networks, potentially including programming networks that are difficult to measure by cs~nv~.tiÃsnal means.
The system of the present ir~~ention preferably implements a reporting system by which individual ~i I EDs 1002 report back to the headend 1004 what asset or assets were delivered at the UED 1002 and, optionally, to whom (in terms of audience classification).
Additionally, the reports may indicate where (o~, what programming channel.) the asset was delivered and how much (if any) of the asset was cons~ed. Such reports 1018 may be provided by all participating UEDs 1002 or by a statistical sampling thereof. These reports 1018 may be generated on a brea:k¾byabreal~ basis, periodically (e.g., every 15 minutes) or may be aggregated prior to transmission to the headend 1004.
Reports may be transmitted soon after delivery of the assets at issue or may be accumulated, e.go, for transmission at a time of day where messaging bandwidth is more available.
Moreover, ~uc~ reporting may be coordinated as between the UEDs 1 002 so as to spread the messaging load due to reporting.
In any case, the reports 1018 can be used to provide billing information 1020 to the T&B system 1006 for val~iing the delivery of the vaiious asset options.
For example, the billing info:tmation 1020 can be used by t1ie T&B sys~~in 1 006 to d~tennine how large an a-Lidiefl'ice received each optioai and how well that audience matc1~~d the target audience. For example, as noted above, a fit score may be generated for particular asset options based on a comparison of the audience classification. to the target audience. This score may be on any scale, e.g., l -1 00. Goodness of fit may be detennined based on tll-is raw score or based on characterization of this score such as "excellent,"
"good," etc.
Again, this may depend on how well an individual audience classification parameter of a user matches a corresponding target audience parameter and/or how many of the target audience parameters are matched by the user's audience classification parameters. This ln~ormatl~ti may in tuni be provided to the asset provider, at least in an aggregated ~orrn.
In this mannerp the network operator can bill based on guaranteed delivery of targeted messages or scale the billing rate (or increase del~~~) based oii goodness of fit as well as audience size. 'nic reports (and/or votes) 1018 can also provide a quick and detailed measurement of user distribtition over the network that can be used to accurately gauge ratings, share, demographics of audiences and the like. Moreover, this information can be used to provide future audience efiti~iatlon information 1022, fi~~ exainple, to estimate the total target universe based on audience classification parameters.
It will thus be appreciated that the ls-resent invention allows a network operator such as an MSO to sell asset delivery under the conventioflial asset delivery (time-slot) buy ~~radlgni or under the new commercial impression ~aradigrn or both. For exaTnlsle, a particular MSO may choose to sell asset delivery space for the major networks (or for these networks during prime time) under the old timeLLslot buy paradigm wbÃle using the cÃ~inmercial i~ipressic~~ paradigm to aggregate users over multiple low market share networks. Another MSO may choose to retain the basic time-slot buy paradigm while accommodating asset providers who may wisl~ to fill a given slot with multiple options targeted to different demographics. Another MSO may choose to retain the basic time-slot buy paradigq-n during prime time across all nehvorks while using the targeted impressic~~~ paradigm to aggregate users at other times of the day. The targeted impression parada~n may be used by such MSOs ori1y 1or this limited purpose.
Figure 11 is a flow cha.rt illustrating an associated process 1100. An asset.
provider (or agent thereol) can initiate the illustTatezi process 1100 by accessing (1102) a contracting lslatf'~rm as will, be described below. Alternatively, an asset provider ~aii work with the sales department or other persa~iinel of a system operator or other party who accesses such a platfoT.T.n. As a still further altemative, an automated buying system may be employed to interface with such a platfc~in via a systerai-to-systern interface. This plat.~~n-n may provide a graphical user zii:~~~~~~ by wlzich an asset provider ~aii design a dissemination strategv (e.go, an ad campaign) and eiiter into a corresponding contract for dissemination of an asset. The asset pro-.rider can then use the interface to select (1104) to execute either a timenslot buy strategy or a targeted impression buy strategy.
In the case of a ~iinemsl~~ buy strategy, the asset provider can then use the user interface to specify (1106) a network and time-slot or other program parameter identifying the desired air times and frequeaicy for delivery of the asset. l'hus, for example, an asset provider may elect to air the asset in connection wgth specifically identified programs believed to have an appropriate audience. In addition, the asset provider may specify that the asset is to appear during the first break or during multiple breaks during the prÃ~graan.
The asset provider may further specify that the asset is to be, for exaniple, aired d~ring the first spot within the break, the last spot within the break or otherwise designate the specific asset delivery slot.

Once the time ~slols for the asset have thus been specified, the MSO causes the asset to be ein1~edded (1108) into the specified ~~~~~imirig channel asset stream. The asset is then available to be coiisumed by all tisers of the programming channel. 'I'he MSO theai bills (111 0) the asset provider, typically based on associated ratings informution. For ~~~inple, ~~~e billing rate a~iay be established in advance based on previous rating information for the program in question, or the best available ratings irafoTTnation, for the particular airing of the program may be used to bill the asset provider.
It will thus be appreciated that the conventional timeasla~t buy paradigm is limited to delivery to all users for a particular timeTslt~t on a pafticular network and does liat allow for targeti~~~ of particular users of a given network or targeting users distribu~~d over multiple networks in a single buy.
In the case of targeted impression buys, the asset provider can use the user interface as described in more detail below to specify (1112) audience classification and other dissemination paranieters. hi the case of audience classification parameters, the asset provider may specify the gender, age range, income rwige, geographical location, lifestyle interest or other information of a targeted audicnce. The add-itiorf,al dissemination parameters may relate to delivery time, frequency, audience size, or any other inform.atir~n useful. to define a target audience. Combinations of parameters may also be specified. For example, an asset provider inay specify an audience size of 100,000 in a par.t~cular demographic group and further specify that the asset is not delivered to any user who has already received the asset a predetermined number of ~imes.
Based on this infomiation, the targeted asset s;rs~ern of the present invention is operative to target appropriate users. For example, this may involve targeting otily selected users of a major network. Additionally or al~ematively, this may involve aggregating (1114) users across multiple networks to satisfy the audience specifications.

For example, selected users from multiple programming channels may receive the asset Witbin a designated time period in order to provide ~audgence ol`the desired size, where the audience is composed of users matcliang the desired audience classification. The user interface preferably estimates the target universe based on the audience classification and disseaninatlon parameters such that the asset provider receives an indication.
of the likely a-Lidience size.

The aggregation system may also be used to do time of day buys. For example, an asset provider could specify audience classification parameters for a target aLidien~e and t.'uzffier specify a time and channel for ai-ri~~~ of the asset. lJl;Ds tuned to that channel can then select the assÃ:t. based on the voting process as described herein. Also, asset providers may designate audience classi~cati~~~ parameters and a run time or time range, but not the programming channel. In this manner, significant flexibility is enabled for designing a dissemination strategy. It is also possible for a network operator to disable some of these strategy options, eeg., for business reasons.
Based on this input inf~nnation, the targeted asset systern of the present inventioii is operative to provide the asset as an option during one or more timeMslots of one or more breaks. In the case of spot optimization, multiple asset options may be disseminated together with infonnat.ion identifying the target audience so that the most appropriate asset can be delivered at individual UEDs. In the case of audieaice aggregation, the asset may be provided as an option in connection With. multiple breaks on multiple programminc, channels. The systeni then receives and processes (111.8) reports regarding actual deli~ery of the asset by UEDs and infor~natlon indicating how well the actual audience fit the classification ~arainoters of the target audience. The asset larovid~.~ can then be billed (1120) based on guaranteed delivery and goodness of fit based on actual report i~~~~nation.. :It will tlius be appreciated that a new asset delivery paradigm is defined by wliieh assets are targeted to specific users rather than being associated with particular programs. This enables both better targeting of individual users for a ~~~~~~
program and improved reach to target users on low- share networks.
From the foregoing, it will be appreciated that various steps in the messaging s~q-uens~e are directed to matching assets to users based on classification parameters, allowing for goodness of fit determinatioiis based oii sucli matcl-dng or otherwise depending on communicating audience clafisif~catia~ii information across the network. It is preferable to implei~ent such messaging in a manner that is respectful of user privacy concems and relevant regulatory regimes.
MÃ,ach of the discussion above has referenced audience classification ~~at'neters as relating to individuals as opposed to liouseiiÃsiÃlse Methods for identifying audience classification parameters are set fortii in co-pending U.S. Application No.
11/332,771, entitled, sbVOI'iNG AND HEADEND UNSEWFiON," the contents of which are incorporated herein by reference. In a first implementation, logic associated with the UED uses probabilistic modeling, fuzzy logic andr`Ã~r m~cbirÃe iÃ~aming to pjogressively estimate the audience classification parameter values of a current user or users based on the ciick. st~earn. This process inay optionally be supplemental based on stored information (preferably free of sensitive infonnat.ion) conceming the household that may, for ~~anipie, affect probabilities associated with particular inputs. In this ~nanner, each user input event (Wbich involves one or more items of change of status an(i6or duration information) can be used to update a current estimate of the audience classification parameters based on associated probability values. The fuzzy logic may involve fuzzy data sets and probabilistic algorithms that accommodate estimations based on inputs of varying and limited predictive value.
In a second implementation, the ciick. stream is modeled as an incomplete or noisy signal that can be processed to obtain audience classification parameter information.
More specifically, a series of clicks over time or associated information can be viewed as a time-based signal. This input signal is assumed to reflect a desi-reÃi signature or pattem that can be crsffeLated to audience classification parameters. However, the signal is assumed to be incomplete or noisy - a common ~robleTn in signal processing.
Accordingly, filtering techniques are employed to estimate the "true" signal from the i~iput st~~~i aniassociated algorithms co-rTeiate that signal to the desired audience classification inf~nnation. For example, a nonlinear adaptive filter may be used in this regard.
One of the audience classifications that may be used fbr targeting is lÃscatioli.
Specifically, aii asset provider may wish to target Ã~illy users within a defined geographic zone (e.g., proximate to a business outlet) or may wish to target ciifff-Tent assets to different geÃ~grapbic zones (e.g., targeting different car ads to users having differeÃit supposed income levels ba-sed on location). In certain impi~~entatioÃis, the present invegition determines the location of a particular UED and uses the location ilii`onnation to target assets to the particular tJED . It will be appreciated that an ~nd~cation. of the location of a UED contains infonnation that may be c.~~~ide-ret~ sensitive.
The present invention also creates, extracts andlor receives the location infonnation in a manner that addresses these pii~~cy ~~~cems. This may also be accomplished by ,~eneaaliziiig or otherwise filtering out sensitive info-ranation fron~ the location infonnation sent across the nehvork. This may be accomplished by providinig filtering or sorting features at the UED
or at tY~~ headend. For example, information that may be useful in the reporting process (i.e. to d~~ennan~ the iiumb~~ of successful deliveries within a specified location zone) may be sent upstream with little or no sensitive in~~~rmation included.
Additionally, su~b location infonnation can be generalized so as to not be personally id~litifiable. For example, all users on a given block or within another geographic zone (such as associated with a zip plus 2 area) may be associated with the same location. identifier (e.g., a. centroid for the zone).
Similarly, it is often desired to associate tags with asset selections. Such tags are additional infonnation that is superimposed on or appended to such assets. For Ã~xarnple, a tag may provide infonnation regarding a local store or other business location at the conclusion of an asset that is distributed on a broader basis. Conventionally, ~-uck~ tags have been appeiided to ads prior to insertion at the ~ea(i.end and have been.
limited to coarse targeting. In accordance with the present invention, tags may be targeted to users in particular zones, locations or areas, such as neighborhoods. 'ra,~s may also be targeted based on other audience ~lafisz~acatioii. parameters fiuch as age, gender, income level, etc.
For example, tags at the end of a department store ad may advertise specials on par-ticular items of interest to particular demo,~~~hics. Specifically, a tag may be included in an asset flotilla and conditionally inserted based on logic contained within the UED 1 10 1.
Tliuss the tags are separate units that can be targeted like other assets, however, with conditional logic such that they are associated with the corresponding asset.
Targeting may also be ~~~~cmented based on marketing labels. Specifically, the headend may acquire information or marketing labels regarding a user or household from a variety of sources. These marketing labels may indicate that a user buys expensive cars, is a male 18v 24 years old, or other in~orinat~on of poteaitaal interest to an asset provider.
In some cases, this infonnation may be similar to the audience classification parameters, though it may optionally be static (not varying as television users change) and based on hard data (as opposed to being ~unnised based on viewing pa~ems or the like)o hi other cases, the marketing labels may be more specific or othenvise different than the audience classification. In any event, the headend may inform the LJED as to what kind of user/household it is in temis of marketing labels. An asset provider can theai target an asset based on the ~~rketin~ labels and the asset will be delivered. by UEDs where targeting matclieso This can be used in audience aggregation and spot optimization contexts.
I'lius, the targeted asset system of the present iiiv~iition allows for targeting of assets in a broadcast netWÃ~rk. based on any relevant audience classi~catiÃsns whether Ãiet~miined based on user inputs such as a click stream, based on marketing labels or other informatiÃs~ pushed to the customer premises equipment, based on demographic or other information stored or processed at the headend, or based on canibinations of the above or other information. In this ~~gard, it is therefore possible to use, in the context of a broadcast network-, targeting concepts that have previously bee.n limited to other contexts such as direct mail. For example, such targeting may make use of financial information, previous purchase i~~brmation, periodical subscription inf'ormatiÃs~ and the like. Moreover, classification systems developed in other contexts, may be leveraged to eiik~ance the value of targeting achieved in accordance with the present invention.
An overview of the system has thus been provided, including introductory discussions of major components of the systenip w1-tich provides a system context for understanding the operation of those components.

~114 COMPONENT OVERVIEW
A. Measurement and VotBn Generally, signals received from a UED 1 002 are utilized by the present systems and methods for at least three separate applications, which in some instances may also be ~oiyilZined. See Fig. 10. These applications may be terrned measurement, voting and reporting. Reporting is described in more detail below. M:easu~ement relates to the use of the signals to identify the audience size and, optionally, the classification composition of the audience. This information assists in estimating the universe of users available for targeting, including an estimate of the size and composition of an audience that may be aggregated over m-ultiple channels (e.g., including low shaTe channels) to fÃ~rrn a substantial virtual channel. Accordingly, a tar~ete~. asset may be provided for the virtual channel to enhance the number of users who receive the asset. Voting involves the use of sigiials received from UEDs 1012 to provide an asset based on asset perforrnance indications from the UEDs. In any case, assets may be selected and inserted itito one or more transmitted data streams based on signals received from one or more UEDs.
With regard to audience measurement, the tWoTway communication between the headend and UED allows for gathering information which may indicate, at least implicitly, inforrnation regarding audience size and audience classikication composition.
In this regard, individual 'LTE:I~~ may periodically or uport, request provide a signal to the headend indicating, for example, that an individual LJED is active and what channel is currently being displayed by the L1ED. l'his inforrfl}ation, which may be provided in connection with voting, reporting on other messages (e.g., messages dedicated to measurement) can be used to infer audience size and composition. Wholly apart from the targeted asset system, such iniormat~on ir~~~ be useful to support rati~~~s and share infornation or for any other audience measurement objective. Referring briefly to Fig. 7, it is noted that of the available programming channels, four programming channels have the largest individual share of users (e.g., the four major networks). 1-1owever, there are numerous other users in the network albeit in smaller shares of the total on a channe1_by-channel basis. By providing a common set of asset options to the users of two or more of the lar~gramminc, channels having a small market share (or even to users oi'prograirming c1iannelS with large shares), a virtual channel may be created. That is, a common asset option or set of asset options may be provided to an aggregated group from multiple progr~nming channels. ~ii~~ combined, the effectgve market share of a virtual chan-ne1 composed of users from small share channels a~~ay approximate the market share of, for example, one of the four major networks.
Wbile the aggregation of the users of multiple larogramming, channels into a virtual channel allows for pjoviding a common set of asset options to each of the progr~ninflng channels, it will be appreciated that the asset will generally be provided for eaclz individual programming claannei at different times. This is shown in Fig. 12 where two difYerent programming channels (e.g., 1202 and 1.204), which may be coinbined into a virtual channel, have different scheduled breaks 1212, 1214. In this regard, an asset may be provided on the first channel 1202 prior to when the same asset is provided on the second channel 1204. However, this common aSse-t may still be provided within a laredeterrnined time window (e.g., between 7 p.m. and 8 p.m.). In this regard, the asset may be delivered to the aggregated market share represented by the virtual channel (or a subset thereof) witMn defined constraints regarding delivery time.
Al~ematively, the size of sucb. an aggregated audience may be estimated in advance based on previous reporting, ratings and census data, or any other ~er-lxnique. Thus measurement or voting is not necessary to accomplish targeting, though such detailed asset infÃ~rmalirsn is usef-ul.
Actual delivery may be ~en~ed by subses1u~~~~ reporting. As will ~e appreciated, such aggregation allows a network operator to disseaninate assets based on the increased market share of the virtual c1iannel(s) in -relatiÃsn to any one of the subsumed prÃ~~armning channels, as well as allowing an asset provider to more effectively target a current viewing audience.
Ariother application that is supported by signals from UEDs is the provision of targeted assets to current users of one or more channels wlxliin the network, e.g.,1sased on voting. Su.ch an applicati~ii is illustrated in Fig. 13, w1iereY in one arraiigeinenty signals received from UEDs 130 (only one shown) may be utilized to select assets (e.g., a break asset and/or lsrs~grammin~) for at least one programming chamel 1350. In this regard, sucli assets may be iiynanlically selected for insertion into the data s~eani of the progra.rrn-iing channel 1350, for exw-npley during a break or ot1i~r designated time period.
In a further arrangement, unused 1~andwidth. of the network is utilized to provide parallel asset stTeams during a break or designated time period of the targeted channel 1350. In the context of a break, multiple asset channels 1360A4N may be used to provide asset options during a single break, wherein each asset channel 1360A-N may provide options directed to different groups of vie-Wers and%or otlierwise carry di~~~~~~t assets (e.g., users having similar audience classification parameters may receive different assets due to a desired s~~~~encing of packaged assets as discussed below).
In such an arrangement, the I1 E:D 1310 ins.y be operative to select between altemate asset channels 1360A-N based on the signals from the UED 1360. 1~
ad(i.ition to targeted audience aggregs,..~iofli, such a systern may be desirable to enhance revenues or ia~~ac~ I-Z)a~ ~rogra~ining, including large share programming (spot optimization). 'niat is, a single break may be apportioned to two or more different asset providers, or, a single asset provider may lsrr~-vi~e altema~e assets where the a11emate assets target different groups of users. Though discussed herein as being directed to providing different break or interstitial assets to different groups of users, it should be noted that the system may also be utilized to provide different programming assets.

~ n associated asset targeting system implementing a voting process is also illustrated in Fig. 13. The asset targeting system of Fig. 13 has a platform 1304, which includes a stnict~e of the network (ioe.P upstream from the usersfhouseholds) that is operative to communicate with UEDs 1310 (only one shown) within the network.
I'lte illustrated UED 1310 includes a signal processing device 1308, wbich in the present illustxation is embodied in a DS'1'B. Generally, the platfomi 1304 is operative to cÃagnmuni.cate with the UED 1310 via a network interface 1340. In order to provide parallel asset channels 1 ~60A-N during a break of a programming c1iannelj e.g., channel 1350, the platform 1304 is in communication Witli one or more of the following components: a schedule database 1320, an available asset option database 1322, voting database 1324, a flotilla construc:tor 1326, a channel arbitTator 1328, and an inserter 1330.
O1`notep the listed components 132041330 do not have to be located at a ~onimon network location., That is, the various ~omponeaits o1'the platfonn 1304 may be distributed over separate locations within the network and ~iav be interconnected by any appropriate cora~iunacation interfaces.
Generally, the schedule database 1320 includes infonnation regarding the timing of breaks for one or more programming channels, the asset option database 1 322 includes available asset metadata iÃ1entifying the asset and targeted audience classification parameters, and the voting database 1 324 includes N"oting information obtained from one or more LJEDs for use in targeting assets. '1 he actual assets are generally included in a separate database (not shown). The flotilla c.onstsuctÃsr 1326 is utilized to populate a break of a programming channel aiidlor asset channels 1360A-N with selected assets.
'l'h~ channel arbitrator 1328 is utilized to arbitrate the use of limited bandwidth (e.g., available asset channels 1 360A.-N) w1~~ii a conflict arises between breaks of hvo or more su~poflteÃi prÃ~~~mning char.anels. Finally, the inserter 1330 is utilized to insert selected assets or targeted assets into an asset stream (e.g., of a progranuning chaimel 1350 and/or one or more asset claarmels 1360A-N) prior to transmitting the stream across the network interface 1340. As will be discussed herein, the system is operative to provide asset channels 1. 360.A,rN to support asset options for breaks of multiple programming channels within the network.
In order to provide asset cla,a.amels 1 ~60A-N for one or more progran1ning channels, the timing of the 13~eaks on the relevant progranimia~g c-hannels is determined.
For instance, Fig. 12 illustrates three progr~imin~ ~hannels that may be provided by the network operator to a household via a nehvork interface. As will be appreciated, many more channels may also be provided. The channels 1202, 1204 and 1206 ca~i-npris.e three progra.~~ing streams for which ~~ge-ted assets are provided. Users may switch between eacb of these channels 1202, 1204 and 1.206 (and generally many more) to select between p~~granuni~g options. Each channel 1202, 1204 and 1206 includes a break 1212, and 1216, respectively, dunng the programming period shown. During breaks 1212-one or more asset spots are typically available. '1'hat is, a sequence of shorter assets may be used to fill the 90nsecond break. For exaniple, two, three or 1`~u-r spots may be defined on a single channel for a single break. Different numbers of spots or avails may be provided for the same break on different channels and a different number of c1iaiuiels may be used for different portions of the break.
In order to provide notice of upcoming breaks or insertion opportunities within a.
break, programming streams oflen include a cue tone signal 1230 (or a cue message in digital networks) a predetemiineÃ1 time before the beginning of each break or insertion opportunity. These cue tone signals 1230 have historically been utilized to allow local asset providers to lnsert localized assets into a network feed. Further, various channels may provide window start times and window end times during which one or more 1~~eak-.s will occur. I"liese start and end times define an avail window. Again, this inforinatlon has historically been provided to allow local asset providers to insert local assets into a broadcast stream. This information may also be utilized by the targeted asset sysl~in to Ã1~~~ine when a break will occur during programming. Accordingly, the systern may be operative to moni'tor progransming channels, e.g., 1202, 1204 and 1206, for cue tone signals 1230 as well as Ã~~taaii and store information regar(ling window start and. end times (e.g., in. the schedule database 1320). The available window information may be received from the T&B systerrf, and may be manually entered.
R~~erra~~g again to Fig. 13, the use of signals from the UED 13 10 may allow for providing assets that are tailored to cu~ent users or otherwise for providing different assets to different groups of users. [n this regard, aii asset that has targeting lsaranielers that match the classification parameters of the greatest n-un-iber of users may be provided within the broadcast st esm. of a supported p~~granim1ng chamel 1350 during a break. It is noted that the most appropriate asset may thereby be provided to analog or otherwise nonparticipating users (assuming the voters are representative of the relevant user unaverse)p yieldinc, a degree of targeting even for thein. Moreover, some targeting benefit can be achieved for a large number ol'progr~~ing chatmelsp even ~hamels that may not be supported by asset channels with respect to a given break., Altemative1y or additionally, different assets may be provided oai the asset channels 1360AyN during the break of a prograraiming chaiuie1. DiniÃ~g a break where asset channels 13 60A-N are available, a UED 13 10 of a particular hoiiseholcl may, based on a determination inzlrlemen.xed at the UED 13 1 03 switcli to one of the asset channels 1360A-'N that contains appropriate assets. Accordingly, such assets of the asset channel 1360A-N may ~e displayel. during the break. During the break, the UED 13 10 may stay on one asset channel 1360AmN (in the case of a break with multiple spots in sequence) or may navigate thrs~ugb the break selecting the most appropriate assets. After the break, the UED 1310 may sWitcli. back to the original programming channel (if iiecessary), This switching may occur seamlessly from the point of view of a user. In this regard, different assets may be provided to different users during the same break. As will be appreciated, this allows asset providers to target different groups dining the same break.
Further it allows for a network operator to market a single spot to two different asset providers on an apportioned basis (or allow a single asset provider to fill a siiigle spot with multiple asset options). Each asset provider may, for ~xarnlsle, thereby pay for an audience that better matches its target.
Fig. 14 ill-ustrates the use of four asset channels 1401-1404 for providing a flotilla of assets d-Lirirag a bre-ac 1410 of a programming channel 1400. As shown, on each asset channel 1401-1404, the break 1410 may be separated into one or more asset slots that may have different durations. However, in the case of Fig, 14, the start and end times of the asset sets A-C, D-E, F-H and I-K carried by the asset channels 1401-1404 are aligned with the start and end times of the break 1410. T;acYi of the asset channels 1401 r 14Q4 may carry an asset that is targeted to a specific audience classification of the users of the targeted channel 1400 or the users of additional programming channels having a break aligned with the break 1410 of the prog,r~~in~ channel 1400.
It should be noted that flotillas need not be rectangular as sh~~~~ in Fig.
14, That is, due to conflicts between breaks or the intermittent availability of certain asset channels as discussed above, the total number of asset channels used to support a given programming channel may change during a. break. Each asset channel 1401 m 1.404 includes a different combination of assets A-K that ma~ ~~~ targeted to different viewers of the channel 1400 during a given break 1410. Collectlvely, the assets A-K
carried by the asset channels 1401-1404 define a flotilla 1405 that includes assets that may be targeted to different groups of users. The inost appropriate assets for a given user may be on diff~~~iit ones of the chan-nels 1. 401-1 ~~4 at different times during the break 1410.
These can be delivered to the user by channel hopping during the break with due consideration given to the fact that spots on different c1~aiinels 1 401 -1.4~~4 may not have the same start and end times. Selection of assets to fill a break of a programming channel, or to fill the available spots within each asset c1iannel of a flotilla may be based on votes of users of the ~~ogranunin~ chaiinel. '17hat is, assets may be selected by the flotilla constructor 1326 (See Fig. 13) in response to signals received 'Tom UEDs 13 10 withiii the network. Sucb. selection may be ~erfo~ed as set forth in co-pending U.S.
Application No. 1.1/332,771, which is incorporated by reference herein.
It is also desirable that each customer premises ~quipm~iix device be able to navigate across a break selecting assets that are appropriate for the current user. For example, a flotilla may include a number of col-umns correspoflidea}t to a sequence of asset spots for a break. If one column included all assets directed to children, non-children users would be left without an. ap~rolsnate asset option for that spot. Thus, options for avoiding such situations include making sure that a widely targeted asset is available in each column or time period, or that the union of the subsets defined by the targeting constraints for each asset in a colunin or time period represents the largest possible subset of the universe oI`aiserso Of course, this may conflict with other flotilla construction goals and an optimal solution ~iay need to be arbitrated. In additioTz, where an issue arises as to which assets to inel-ude in a flotilla, the identity of the relevant asset providers may be considered (e.g., a larger volume asset provider or an asset provider. who lias paid li~r a higher level of service may be given preference).
To enable the UED to switch to a designated asset cham}el for a break (or, for certain implementations, between asset options witliiri the flotilla during a break) metadata may be provided in ~~nn~ctia~ii witli each asset channel(s) and/~~
programmia~~
channel(s). As will be appreciated, each individual asset channel is a portion of an asset stream having a predetermined bandwidth. These asset channels may be further broken into in-band and out-of-band portions. Gr~~ra11y, the in-band portion of the signal supports the delivery of an asset stream (e.g., video). Triggers may be transmitted via the outmof 1~aiid portion of a channel. Furtlier, such out-of-band portions of the bandW1~~b may be utilized for the delivery of the asset option list as well as a return path for use in collecting votes and reporting inforÃnation from the UED. More ~en~ally, it Will. be appreciated that in the various cases referenced herefnwl~~~e messaging occurs between the UED and a ~~twork. platform, any appropriate messaging channels may be used including separate [P or telephony channels.
Based on the metadata, the UED may select individual assets or asset sets depending on the implementation. `I'hus, in certain implementations, the UED
may select an asset for the first tim.e-slot of a break that best cozrespa~iids to the audience classification of the current user. 'I'hls process may be repeated for each t1me-slot within a break. Altematively, an asset flotilla may include a single metadata se-t for eacb. asset chaiinel and the UED may slrnply select one asset c1ianiiel for an ~iitire break.
A1t~matively, asset options may be provided via a fonvardTars.dTstore architecture in the case of UEDs with substantial storage resources, e.g., DV.Rs. ln this regard, ati asset may be inserted into a slesagnated bandwidth s~~ne~t and downloaded via the network interface to the storage of the UED. Accordingly, the LJE~ may then selectively insert the asset from the storage into a subsequent break. Further, in this archi~~~tureo the assets of the stored options and associated metadata may laielu~e an expiration time.
Assets may be discarded (e.g., deleted.) upon expiration regardless of whether they have been delivered. In this architecture, it will ~e appreciated that the transmission of assets does not have a realmtime componeait, so the available bandwgdtb. may vary during transmission. Moreover, a thirty second asset may be transmitted in five seconds or over thirty minutes. ~~e available assets may be broadcast to all UEDs with individual UEDs onlv storing appropriate assets. In addition, due to storage 1~~~~tations, a UE,D may delete an asset of iii:~erest and reLLrecwrd it later.
In contrast, in the asset channel architecture, the flotilla is transmitted in synchronization with the associated break and requires little or no storage at the ~jED. In eitlier case, once an asset from the storage or flotilla is displayed, each UED mav provide an asset delivery tiox~~catlon (ADN) to the network platform indicating that the particular asset was delivered. The platfonn may then provide aggregated or compiled information regarding the total number of users that received a given asset to a billing platform. Accordingly, individual asset providers may be billed in accordance with how many users received a given asset.

B. Dynamic ~chedul As noted above, the system allows for dynamically inserting assets in support of one or more programming channels based on current network coiiditions. That is, assets may be selected for programming channels in. view of current network conditions as opposed to being selected ahead of time based on expected netwa~rk, conditiotis. Such a process m~~ ens-ure that high value air time is populated with appropriate assets. For instance, where current network conditions may indicate that an audience is larger than expected -f~r a current p~~~~~nmia~~ period, higher value assets may be utilized to populate breaks. Such conditions may exist w1ieii, for example, prograr.nmfng with liigli asset delivery value and a large expected audience extends beyond a predetermined ~ro~ammi~~ period into a subs~q-uent programming period witli low asset delivery value (e.g,5 a sporting event goes into overtime). Previously, assets directed to the subsequent low value progranuning period might be aired to the larger thaii expected viewing audience based on their pre-scbeduled delive-ry times resulting in reduced revenue opportunities. The present system allows for dynamic (eog., austninmti~~) asset scheduling or, at least, overriding pre-scheduled delivery based on changing network conditions.
As noted, signals from the individual UIEDs may be utilized ~~~ targeted asset system purposes. However, it will be appreciated thatWhile it is possible to receive vote signals ftom each IJED in a network, such full network 'polling' may result in large bandwidth requirements. In one altemate implementation, statistical sainpling is utilized to reduce the bandwidth requirements between the network and the UEDs. As will be appreciated, sampling of a statistically sigiii~cant wid rel~vant poTtion of the TJEDs will provide a useful representation of the channels currently being used as well as a useful representation of the most appropriate assets for the users using those ch.aimels.
In order to provide statistical sampling for the network, a sub-set of less than all of the UEDs may provide signals to the network platform. For instance, in a first arrangement, each UED may include a ran(i.om number generator. Periodically, such a random number generator ~ia5y generate an output. li~ this output meets a pre-d~~ermined criteria (e.g., a number ending with 5), the UED may provide a signal to the network in relation to an option list. AlÃ:ematively, the platfon-n may be operative to randomly select a subset of'1J'EDs to receive a request for iraf~rmati~ii, In any case, it is preferable that the subset of UEDs be large enough. in comparison to the total number of UEDs to provide a statistically accurate overview of current network conditions. However, where a fully representative sampling is not available, attendant uncertainties can be addressed through business rules, e.g., providing a reÃla~ced. price or ,~cater dissemination to account for the u~~~~rtainty.

As noted, a network operator initially provides an asset option list (e.g., list 1010 of Fig. 10) to at least the UE: s within the network that will vote oti assets f-rom the list.
Generally, the asset option list includes a list of available assets for one or more upcoming breaks. In this regard, it will be ap~~~~iated that a platfrsrTn within the network iraa~ be operative to obtain sched-Lil~ infonnatican for all programming channels that have been identified to be supported by targeted assets. 17he platfonn may then use the schedule information to communicate with UEDs over the network interface p fior to a break. In particular, the platform may be operative to provide the asset option list to UEDs, for example, periodically.

C. Emorti It would be possible to implemeflit the targeted asset svstem of the present invention without receiving reports from IJEDs indicating wbich assets, from among the asset options, were delivered to the user(s). That is, altl~ougb. there ~ould be consid~rable uncertainty as to what assets were delivered to whom, assets could be priced based on what can be inferred regarding current network conditions due to the voting process.
Such pricing mav be improved. in certain respects in relation to ratings or share-based.
pricing under the conventional asset delivery paradigm. Altemativel,} pncing may be based entirely on demographic rating information such as Nielsen data together Witl~ a record of asset insertion to build an estimate of the number of users who ~~ceived an asset. For ~~aniple, this may work in connectaoii with programming channels that have good rating informat~one However, in connection Witki the UED selection model, it may be desirable to obtain report inforinatit~~ concerning actual delivery of assets. That is, because the asset selection occurs at the T-TED (in either a forward-and.wstore or synchronized transmission architectuire) improved certainty ~~gardiiig the size and audience classification values for actual delivery of assets can be ea~aiiced by way of a reporting process.
"1lie present invention provides an appropriate reporting process and in this regard provides a mechanism for using such report infonnation to enable billing based on guaranteed delivery andf~r a goodness of fit of the actual audience to the target audience. In addition to improving the quality of billing inf~nnatiofli and information available for analysis of asset effectiveness and return on inves~ent, this reporting inforrnaIion provides for near real time (in some r~orting implementations) audience measurement with a high degree of accuracy. In this regard, the reporting may be preferred over voting as a measurement tool because reports provide a positive, after-t1~e-facx indication of actual audience size.
Accordingly, such information may allow for i~~~~~~ed ratings arfld. share data. For exaziples such data may be licensed to networks or ratings measurernea~~
entities.
Fig. 15 illustrates a reporting system 1 500 in accordance with the present invention. The reporting system 1500 is operative to allow at, least some users of a participating user group, ~~~~~rally identified by ~~~~~~n. c~ numeral l. ~~~~
to report actual asset delivery. In the illustrated imlslea~entation, such report inI'oranaIion is transmitted to a network platform such as a headencl 1504. The report infonnation may be further processed by an operations ceiiter 1 506 a:aid a traffic and billing systein 1508.
More specifically, report inl'ornation is gÃ~~~~~~ed by individual UEDs 1513 each of which includes a report processing module 1516, an asset selector module 1518 and a user monitoring module 1520. 'The user monitoring module 1520 monitors inplits from a current user and analyzes the inputs to sl~~errnine putative audience classification parameter values for the user. 'I'hus, for example, module 1 520 may analyze a click stream from a rexn~~e control together with information useful for matching a lsattem of that click stream to probable audience classification parameter values.
These classification parameters may then be used by the asset selector module 1518 to select. an asset or asset sequence from available asset options. 'I'li-us, as deficnbed above, multiple asset s~~~ences may be available on the programming channel and separate asset channels. Metadata disseminated with or in adva:aice of these assets may identify a target audience for t1~e assets in terms of audience classification paranieter values. Accordingly, the module 1518 can select an asset from the available options for delivery to the user (s) by matching putative audience classification parameter values of the user to target audience classification ~~~~~~r values of the asset opIions. Once an appropriate asset option has been identified, delivery is executed by switching to the corresponding asset channel (or remaining on the programming chaiinel) as appropriate.
The report processing module 1516 is operative to report to the headend 1504 inforTnation regarding assets actually delivered and in some implementations, ~enai~l audience classification parameter values of the user (s) to wliom the asset was delivered.
Accordingly, in sucb implementations, the report processing module 1516 receives asset delivery information from module 1518 and putative audience classification parameter information for the user (s) from the user monitoring module 1520. This infionnation is used to populate various fields of a report file 15 10. In other implementations, audience classification information is not included in the report 151.2, However, it may be presumed that the asset was delivered to a user or users matching the tax,~~~
~ara~~~ters.
Moreover, such a presumption ~-nay be supported by a goodness of fit ~~~me-ter included in the rqporta Thus, audience classification infonnation may be inferred even where the ~eport is devoid of sensitive information.
The repoTt files pass througb the headend 1504 and are processed by an operations center 1506. "1'1ie operations center 1506 is operative to perform a number of functions including processiiig report inf~nnatic~n for submission to billing and diagnostic functions as noted above. ~~~ operations center 1506 then forwards the processed report infarTnation to the traffic and billing system 1508. The traffic and billing system 1508 uses the processed report infÃ~miation to provide measurement infoTTnatisan to asset ~~o-viders with respect to delivered assets, to assign appropriate billing values for delivered assets, and to estimate the target universe in connection with developing new asset delivery contracts.
In order to reduce the bandwidth ~equirenients associated with repoiting, a statistical reporting process may be impl~ented similar to the statistical votinc, process described above. In particular, rather thara having all UEDs report delivery with respect to all breaks, it may be desirable to obtain reports from a statistical sampling of the audience 1502. For example, the UED of each user may include a random nuanb~
generator to generate a number in conyzection witl~ eac1i reporting opportunity.
Associated logic may be configured such that the UED will only transmit a report file w1~~~ certain numbers are generated, e.g., numbers ending with the digit "Y' Alternatively, the I~ED may generate reports onlv u-pon interrogation by the headend 1504 or the headend 1504 may be configured to interrogate only a sanililing of the audience 1502. Such statistical r(Torting is graphically depicted in Fig. 15 where users selected to report Witb respect to a given reporting opportunity are associated with solid line links and deselected users are associated with a broken line links.
Moreover, reporting may be batched sueb that all reports for a tiin~ ~eliodY e.g., 24 hours or seven days, may be collected in a single report transmission. Such transmissions may be timed, for example, to coincide with low messaging traffic time periods of the netwcark. Also, the reports from different 1JEDs may be spread over tim.e.
Billing parameters and goodness of fit inf~rrnation may then be determined based on the report info-rmation, The billing parameters will generally include infonnation regarding the size of the audience to whom an asset was delivered. '1'be goodness of fit infonnation relates to how well the actual a-ucii~~~e matched the target audience of the asset provider. In this regard, a preniiuin niav be extracted where the fit is good or a di~coa~~it or credit may be applied, or over delivery may be provided where the fit was not as good. Based on this information, the T&B system can then generate billing records. It will be appreciated that such billing reflects guaranteed delivery of targeted impressions with ca~i-n~ensation for less than optimal delivery.
As noted above, a platform and associated gaphica1 user interface may be provided for receiving asset contract gnfarinationõ As will be described in more detail below, asset providers can use this interface to specify ad campaign information including targeting c~~ena such as geographic infornation, demogaafabic infs~~nation, nm-time infcnnation, ruai frequency information, run sequence infonnation and other infonnation that defines asset delivery constraints. Similarly, constraint information may be provided from other Sources. 171riS contract information may also include certain pricing information includi~~ pricing parameters related to goodness of fit. Moreover, in accordance with the present i~ivention, report infonnation can be utilized as described above for purposes of traffic and billing. All of this requires a degree of integration between the T&B system, w11#.ch may be a conventional product developed in the context of t1~~ conventional asset delivery paradigm, and the targeted asset delivery systean of the prescait invention, which allows for implementation of a novel asset delivery paradigm.
Among other things, this integration requires appropriate configtiration of the T&B sy5tein, appropriate configuration of the targeted asset delivery system, and a definition of an appropriate messaging protocol and messaging fields for ~raiisf~~ of information between the T&B system and the targeted asset delivery 5~~tem.e I'Vitl~
respect to the T&B system, the system may be configured to recognize new fields of traffic and billing data related to targeted as~~t, delivery. These fields may be associated with: the use of reporting data, as contrasted to ratings or share data, to d~~ennine bill~g values; the use of goodness of fit parameters to determine billing ~~rameters;
and the use of report information in estimating the target universe for subsequent broadcasts.

A~corclir~glys the T&B system is configured to recognize a vanety of fields in this regard and execute associated logic for calculating billing parameters in accordance with asset delivery contracts.
The targeted asset system receives a variety of asset contract in~onnation via a defined graphical user iaiterface. 'Tliis asset contract infs~nnatic~~ may set various constraints related to the target audieiice, goo(i.ness of fit paraineters and the like. In addition, the graphical user interface may be operative to project, in substantially real time, an estimated target universe associated with the defined contract parameters.
Consequently, integration of the targeted asset delivery system with the T&B
system may involve configuring the targeted asset delivery system such that inputs entered via the graphical user interface are mapped to the appropriate fields recognized by the ta~~~~ed asset delivery system. :Cn adtlitioii, such integration may involve recognizing report information forwarded from the targeted asset delivery system for use in estimating the target universe. Generally, the T & B systeni is modified to included logic in this regard for using the information from the targeted asset delivery systein to project ~target universe as a function of various coiitract informatia~~~ entered by the asset provider via graphical user interface.

Various c~nibinations of the alsoveLLdescribed systems and methods may be utilized to provide an auctioning pl~~fon-n 1602 for use in auctioning asset delivery spots (eog., avails). See Fig. 16. As shown, the auctioning platfarin, 1602 is accessible by a pl-urality of asset ~-rovide-rs l. 604AwN. Such access may be provided using, for example, a graphical user iai~~rfacex web access, etc. '11~e auctioning plaxfomi allows asset providers to bid on asset delivery spots on one or more broadcast channels. "1'he auctioning platform 1602 ma:y allow asset p-roviclers to upload content (e.g., assets) to the systeni such that the content inay be inserted into broadcast content. In any case, the auctioning lrlatforTn 1602 is in cÃ~nimunication with a headend 1506 that is operative to iinplemea~~
part of all of the asset targeting systems and methods described above.
Further, the auction platfarTn is in ccarimunication with a traffic and billing system (T &
B System) 1608. '1'he systeni described herein allows auctioning of specific avails in specific programs or at specific times on specific channels and1or atactionir~~ of viewer irrt,pressions. The examples below are assumed to be local spots, that is, that sales are within a particular market area, ~-ai the technique generalizes to regional, national and 1n~ematlonal markets.

In a first arrangement, a single avail ~~ay be auctioned to a single wining asset provider. Initially, lnl'orination regarding an asset delivery spot is provided 1702. See Fig. 17. In this regard, multiple asset providers may bid 1704 ~~i an asset delivery spot.
A wim-ting bidder is then determined 1706. Accordingly, an asset of the winning bidder may be delivered 1708 d-uring the delivery spot. The aue-t~~n mechartism itself can be a sianple highest bidder system. ln another ai-~~~~~inenly the auction mechanism may be in the style of a Vickery auction where the highest bidder wins but pays only a penny or other nominal arnount more than a second highest bidder. Two examples of auctions where a single avail is provided are set forth:
1. 1st position in 11 break on "Larry King Live" oti CNN at 21:00 June 7, 2007 2. 1s` position in 2~ break 1~ehveen 22e00 and 23e00 on CNN J-urfle 1~' 2007 In instances Nvhere the asset to be delivered is already available in the system, an a-uction need Ã~ifly conclude a small amount of time before the break window starts.
When the auction concludes, the winning bidder (a:aid in particular the asset associated with the wi~iig bidder) is communicated to a viewlist composer, which in tum arranges for the asset to be inserted into a broadcast content stream. Such insertion may include replacing the default ad in a content stream, transmitting ttie asset of the winner in 211 separate stream in synchrony wltli the avail and then causing the DS"1`B
to switch to the appropriate asset channel aiid.1~r transmitting instructions to the UED to play a specific asset during the asset delivery spot, wh~.~e the asset tirnej has been previously stored on its hard disk. The system may or may not return asset deLz~ery notifications (ADNs) from the IJE1) signzfyitig that the asset has been delivered.

In the above description, a bidder specifies a price for the specific delivery spot and it is presumed that the bidder has knowledge of one or more characteristics of the audience that will be present. An al~~mative mechanism is to provide audience characteristics such as ratings information along with the description of what is being sold/auctioned. Extending the above hvo ~~ainples:
1. 1st position in 1" break on "1arryKing Live" on CN.T at 21:00 June 71',2006 -the national household rating for di.s pr~ograin is 101 2. l.A posifion i-n 2d break between 22:00 and 23 ):~0 on ~~T June 7'h$ 2006 -last weeks quarter hour rath~gs averaged 0.7 A further variation of the auction mechanism that takes advantage of the extra inforrnation (e.g., ratifligsg etc.) allows bidders to bid using familiar price models for advertising sales, for example, cost per thousand (CPM) ~id cost per point (CPP). in such an arrangement, a bidder may choose to place bids in total cost mode, CPP
mode or CPM mode. To facilitate such conversion, the ratings estimate is presumed to be co-rrect, so that these bids are easily converted fTs~~ one to another. For exaniple, in aii auction mechanism that provides converted bids, instead ofbgdding $150 for a spot, a bidder may bid $13 CPM (which is based on an audience size that is calculated as number of households ~~~ched Ibr a speci-fic operator in the specific market niulxiplied by raxingr`1 00) or a bidder may bid $77 (;PP. These th-ree bids may be equivalent.
In a ftirth~~ arrangenleilt, the winning bidder (e.g., the buyer) pays only for the assets that are a~tua11v delivered 1710. For instance, using ~~~~med ADNs, the actual number of impressions (network users who receive a given asset) may be calculated and the winning bidder may be asked to pay for them proportionally based on xlie original rating. Such a model may be referred to as "guaranteed impressions.'R For example, in a market with 1,000,000 households, all of which are reached by a system operator, a broadcast program is estimated to have a rating of 2.0 (meaning it will reach 20,000 households). If a bidder wins with a bid of ~300 for the spot (which in the other methods described would be bidding $150 per point (in C1DP mode) or $15 per thousand (in C;PM:
mode) then the bidder may expect to get 20,000 impressions verified by ADI~.~s. What they actually pay is $300 * (actual ~~~~ience size / 20,000) This model may require the winning bidder to pay more or less than they originally bid Ibr the spot. To provide the winning bidder some certainty, it may be desirable to cap the overage that they would pay. For instance, it may be agreed in advance that a winning bidder will never pay more thaii, for example, 20% more t1ian their actual. bid ~ount if a bigger audience appears. Further, if the actual audience is within some percentage of the original estimate, for example 5%, then. the winning bidder ~~~~ pay the original estimate. Ratings information may come from an extÃmal source like Nielsen or it may be generated using ADNs or votes returned from UE s, or it could be a combination of such information.

The above di~c-usses placing a single asset into an avail (e.g., asset delivery spot).
Of course, this avail could be used for a spotmoptimi~ed spot with several targeted altematives being supplied during the avail because of targeting peiformed at the UEDs.
That is, an advertiser could bid and buy the spot, and then provide three differently targeted assets to be run in the spot with the UEDs of the network users picking the particular asset for the UED of eac1i user for that I:JED. In suc1i an arrangeniefliI, a multi-spot premium. that is over aiid above the bid price may be charged for such a service.
In a:aiotli~r arrangement, multiple avails may be auctioned to a single winne.r. For instance l . All of the 1 ~ position in l. " breaks on "Lan-y King C,ivex on CNN at 2 1 :00 for the week- of June I~~ ~~ ~~~e I 8'h 2006 (7 Avails) total ratings points 7.7 2. I~ position in 2"d break between 22:00 and 23:00 on C,'~~ for the week starting Jun l. .9ffi 2006 - ~otal average rating from last week 4.9 3. 20 breaks (described here ...) on Network A in the next week. Average rating for this netw-ork is .3, with a ratings guarantee of 6.0 rating points, 4. In the week of Jun 19`h 2006 breaks in the following 30 programs (list follows ~s which total 20 rating points.
In such an arrangement, the auction may need to conclude before the first break of the group, By grouping several programs together the ratings guarantee mechanism may be more easily implemented as the risks associated with a:uciiea~~e variability from day to day can be averaged. As well, by picking a pool of advertisiiig on an unrated network, calculating a likely overall rating, and making a ratings guarantee, becomes less risky.
In another a-rrangement$ as illustrated in Fig. 18, a single avail may be auctioned to multiple wiiiners. That is, as the spot optimization system can provide multiple advertising options at one time, those multiple options for a single asset delivery spot may be sold to a~~-ultiple bidders. Examples of a multiple option single avail auction:
1. 1 `t position in l ~ break on "Larry King Li-ve's on CNN at 2 l :00 ,Iu~~c '7h, 2006, two wimers each getting 50% of the audience 2. 1.S` position in 2d break ~etweeii 22.00 and 23:00 on CNN June 7'112006, three ~~imiers each g~tfi-ng 33.3% ol`the audience Initially, inforr~iation associated witb. the avail is provided I802 to the asset providers. Provision of inf~rynaIion may include providing one or more audience cliaracIeaisI.icse The asset delivery spot is then auctioned 1804 to Ille aSset providers based on two or more characteristics (e.g., a 1/2 audience share, demographics, etc.).
Winning bidders are d~~erminef. 1806. Assets of the winning bidders are inserted 1808 into parallel content streams and delivered 1810 during the asset delivery spot (e.g., simultaneously). ln this regard, a first asset may be delivered to a first portion of a broadcast audience, and a secoizd asset may be delivered to a second portion of the broadcast audience.

As will be appreciated, multiple options for a single avail require either simultaneous synchronized transmission of the assets or playback from local storage. As discussed above, the l..." EDs will pick which asset to show based on, for example, a random number ~eneration. For instwice, using a random number ~er~~ata~r t1l~.~
generates real numbers in the range [0.0,1.0], then given the scenario described in exaniple above, the auel.iezace may be split between two different win-ners.
Of course, the auction changes s-ubtly to a~~~nimodate multiple winners (e.g., two or more).
All of the winners may pay the same amount for their win. Altematives are that the winners all pay the price d~~ennined by the lowest winning bidder, or they pay a penny more than the highest losiiig bidder.

In a further arrangement, the audience for a specific program may be identified by Ãiernographicfi and each of those demographic may be auctioned separately.
This may represent a rating for specific demographic group, rather than a household rating. An example auction would be 1. In 1 " position in 1 ", break on "Larry King Live" on CiNN at 21:00 Jwie 7h5 2006:
gMen 55-1, - raiing 1.2 ~Women 55+ - rating 1.8 - Remaining au(l.ience T rating 1.0 Here, a bidder would bid on one, or more, these demographics, which may each be sold in a separate auction. A bidder may choose to compete for more thai~
one of the demographics, and will likely pay a differing amount for each demographic won.
Note that in this exainple the dern~graphics do not overlap. However, this is not an absolute ~~quiremento as a mechanism for randomly assigning a given demographic grati~
to multiple winners with a randomized delivery may be implemented. Such a mechanism may be used to split overlapping deins~~aphic categories between winning bidders.

This may fiirther be generalized to split the audience of each program auctioned into, for example, the 16 age/gender ranges that Nielsen uses for demographic rating.
Each of these ranges is non-overlapping (the age ranges are 2-11, 12-17, 18-24, 25-34, 35-49, 50-54, 55-64, 65+ and are calculated for both genders). A bidder may compete in separate auctions for each demographic that they are interested in, Note that in many progranis the ratitig for a given category may be zero or nominal, and thus, no auction may take place for such a demographic.
In a farther arrangement, a bidder is allowed to specify aii all-or-nothing bid. That is, their bid is allowed to be conditiotial on winning eacb. of their auctions, or even sorne specified fractioai of their bids. This may be dealt with by detennining a "potential winner'F by deciding if their bid criteria has been met and if not, knocking them out of the auction and elevating the second place bidder in all of the auctions they have been knockeÃ1out of. This style of auction may be ÃmglerflienteÃi in a GUI that would allow the bidder to easily place bids and establis~ various limits across a group of bids.
In another arrangement, niultigle avails may be auctioned to multiple winners.
For instance, Wlien auctioning off a group of similar avails, it may be desirable to allow bidders the opportunity to bid on subsets of the whole group. 1n this kind of auction, the avails may be similar. Consider an auction for basketballs. There are 20 for sale; a bidder can bid for as many as they want. This is easy for a bidder. But an auction for 20 balls where there are baseballs, basketballs, golf balls and. tennis balls presents a problem for the bidders. In this instance, it may be better to run different auctions for different types of lsalls. ExaTraples ol`~nultipl~ avails raiu:ltlple wiiiners auctions:
l . 14 avails in Larry King Live for the week of June 1 8'h. Note that two a-vails per prÃ~gyam are Ã~fferÃd. Bidders may bid on any nurrt,ber of avails. Average rating pÃ~~nts per avail are 1,1. No impression guarantee provided on purchases of less than 7 avails.
2. 42 prime-time avails on OLN for the week of June 18 Ih . Two avails per hour are offered between 7pm and 1 Opm. Bidders must bid for a mlnirnum of 10 avails to get an impression guarantee.
Again the auction model changes to accommodate multiple winners with the high bidder being allocated their share until all slots are used up. Various pricing mechanisms are possible. Altematives are that each winner pays what they bid (per avail), all wimiers pay the same amount per avail that the lowest bidding winner pays, or all winners pay a penny more than the high loser per avail.

IÃt, the saine manner as described when auctioning a single avails to niultipl.e winners, the demographics for the group of programs may be lZrokea~ apart and each group auctioned separately. These individual auctions can be run either as single winner auctions (in which case the programs need not be similar) or they can be run as described above with bidders bidding on portions of deniographics pools (either by impressions or rating points). In this case, it may be desirable that the pr~granis are similar or have similar audiences. In practice, this may mean groups of the same programs or perhaps large groups of programs on specialty n~tworks.

Exarrip1~ a-uctgons where multiple avails are sold by demograohlcs.
1. 56 avails in Larry King Live for the broadcast mon~i of July 2006 broken into the follc~Nvizig deir~o&Taphac: groups:
-Men 55+ ~ total rating p~i-nts 67 -Womera 55+- 4 total r-ating po1lits 10 1 - R.emafi-dng audience -- total rating points 56 A bidder may bid for any n-Larnber of ratings they desire. Fu:rthLT, to facilitate the process, the number of rating points bid for may '~~ ~~ceeded by up to 2 ratings points (i.e. if a bidder bids for 17 poijals, thev may win 1.9 points).
All of the systerras, to the extent that they use ratings information, may get their ratings lnfor~~atir~n firom an extemal source such. as S ;lelsen., An altemative source of ratings ln-fornatlon is for the systeir~ to use ADNs to build up a model for program ratings. Bv monitoring ADNs and the targeting of assets delivered to those audiences it is possible to mak-.e interences about the size and demo,~aphics of audiences.
i'liese inferences can be a~cu~nulat.esi and used to predict program ratings. In another arrangement, a system similar to voting that returns information about the types of people that are currently viewing is used to provide a aeal-time estimate of the audience for ea~~
ad. This information could be -used just-in-time to detertnirxe auction winners.
Users of this system may riot want to manage hundreds of auctions on an auctgon-by-auctior~ basis. Accordingly, an interface that allows ~i asset provider to automate the process of finding appropriate auctions and then bidding on them is provided.
0a~e 3 '0 component of this system is a search mechanism that helps users find auctions that meet the user's various criteria such as laousebold or demographic rating inforrnataon, current bid ~ounxs and historical bid amounts. Another component of this systern is an automatic bidder that automatically submits bids on specific types of avails.
For instance in a system where individual avails are split apart by demographics, the automated bidding system may take bids such as "please bid 1 50 CPM on any ~~en t 8-24 dernographics where the rating is betweeri 0.5 and LO."
The core concept for this mechariism is to integrate an aggregation model with a jaast-in-time auction. 't'he key for an aggregation model is that the asset provider/bidder describes a set of target attributes for consumers that they wish to reach and then the syytern helps thern reach that audience across a group of chan-nels 24 hours a day (or other time fr~e as set forth by the bidder).
A bidder begins the purchase process by usinc, a GUI (or other systeni-to-system interface) to sts~~~t-y the parameters for an aggregated auction offer. The tsaranieterfi for an offer allow the auction system to make automatic bids on behalf of bidders.
The parameters include:
1 o Targeting criteria - many different targeting meebanisms may be ~e(le A
given ad insertion amp1ernentatiÃ~~ ~nay suppÃ~lt only a subset (or a superset) of the following:
s UED classified age and gender g I-J~~ ~lassificÃi household income m Start and end time and date for campaign - Time of day I"unitations 8Day of week limitations ~ Geographic restrictions ¾ Household tags (detcTrF,~ned using UED identifier lists from the head~id that directs the LED to select a particular ad or type of ad) = Network inclusions and exclusions = PrÃ~gm rating inclusions and exclusions - Prograin title word inclusions and exclusions = Keyword searches = C;omirios~~~y codes g Minimum separation 2. Maxirnum impressions - an asset prÃ~~~der specifies a. total number of iinpressions that they want to buy. Once this total is reached the offer is deern fulfilled and aLitÃ~matac bidding stops.

3. Maximum price per impression - an asset provider specifies the maximum amount of money that the automatic bidding system sliould bid per impression.

4. Maximum cost - an asset provider specifies the maxlnium amount of money that the buyer is prepared to pay for the contract. Once this arnount of money has bÃ~~
expended on the camlsaip, the ofTer is deemed fa1.fi1l~ and automatic bidding stops.
5. Pacing - the asset provider may specifsr pacing constraints that specify the maxiTnum am~tmt of money the prc~~ider is willing to pay for a given ~~~~~
peTioÃ1., These c-an be specifiei. &s daily, weekly or monthly pacing ainounts. In any given time pei-1od if the specified total is reached then automatic bidding is su~pend~.Ã~. until the next period staits.
Note that all of the above may be changed at any time, although there may be a delay in implementing some of the. changes. For instance, in a given sys~e.m.
it ma~it tak-e up to 24 hours to make chaiiges to targeting whereas updates to maximum price per impression might take effect nearly instantly. Other changes inight take effect only once per da~ at a given time of day (for instance changes to pacing may take effect at 2 am each momlng). A given. campaign may also be suspended and resun-ied (that is, automatic bidding stops until the campaign is resumed).
Asset providers bid on targeted impressions to be delivered to audiences.
These impressions may be sold by nuu-iÃng an automatic auction before each break occurs on a network for which auctioning insertion is suplsa~~ed, ln general, an asset provider will need to win a number of auctions to satlsfv their impression goals. Each asset provider may enter the auction for each possible avail or asset providers may elect to enter only selected auctions.
One process for implementing the justaln-time automated auction is provided in relation to Figure 19. Initially, the auction lslatl'oain receives 1902 and campaigns from asset lsrovaders. These canipaagns may be received over a considerable period of time andi`~~ on an ongoing basis. On a periodic basis, a list of the targeting constraints for all of the active campaigns is transmitted 1904 to all I_JE:Ds in the system. 17he set of constraints that ~~e transmitted to the UEDs include those constraints that can only be evaluated in the UEDs. Shofl-tly before the break- window for an avail on a given network occurs the system asks UEDs to "vote." At least a statistical sample of UEDs tuned to the network in question submit votes that list one or more, e.g., tll~ complete set of campaigns that the UEDs matches at the moment of the vote. The auctioning platform collates the votes that are received 1906 from the UEDs.
The system may evaluate some of the targeting ~~eria in the 1~eadmend and/or auctioning pl~~~on.n and dexerrnine 1908 that certain campaigns are not eligible to be played even thoue-i some UEDs vote for them (for instance, program rating exclusion might be Ãletennined only in the head-enÃl). Votes for these campaigns are eliminated.
The size of audience for each eligible campaign is estimated from the collated votes and the voting sampling crikeria, The auction system uses the iiifamiatlon from the aud1e~.ce size estimation and the offer parameters to determine 1910 the winner of the auction. A.
price per impression is also deteaTnined if an additional parallel distribution opportunity is available, the-n all votes which include the winning campaign [~om the previous step are eliminated, the remaining votes are recollated and steps 1906 to 1910 are repeated until there are no remaining distributlon. opportunities. In one arrangement, the price per 1mpre ssion f~r all ~.~1"tl~~ winning bids is adjusted (e.g.y based on a Vickery style auction).
Provisional updates to the impression totals, and cost totals for all of the wimii~~
campaigns are accounted for. All of these provisional updates are tracked in a manner that allows thern to be "backed out." When the cue slgnal. arrives t-he set of assets associated with the winning campaigns are distributed 1912 in synchronized parallelism wlth the avail. Each LJED tuned to the channel picks an ad for insertion, and then each UED, or a statistical sample of UEDs, reports whicli of the assets that it delivered to the headed (e.g.} Asset Delivery Notifications or ADNs). The winning bidders may then be charged based on the actual number of impressions that were delivered. 'C"o do this, the actual number of impressions delivered is multiplied by the cost per impression calculated for tMs campaign during the auctionõ The provisional update for each winning campaign asbacl~ed out and the actual impression count and costs are used to update the totals.
'~~~e noted automated auctioning mechanism uses a voting mechanism to estimate the size of an audience. As a UED evaluates all of the UED dependent parameters to d~termln~ a match, each vote provides a very accurate estimate of the campaign matching the UED audience for the lynpending break. However, there are altematl~e mechanisms that could provide an estimate of the size of audlerz~e for a particular carnpai~ for an upcoin1ng break. The accuracy of these meelianisms will depeiid on the set of targeting mechanisms available in the system. Altematlves include:
1. Use external data sources that include television ratings and census ~~~
2. U'se historical AD'Y data to build up a statistical model of ~eNvershap 3. Operate the voting system to periodically survey the system for information about curreiit viewers (as opposed to eligible campaigns). To differentiate this mechanism from ~~~~ we will call this a ~D census"

Alternate Mechanism for:~~~ermining Price As described, the automated atictioni.ng meelianism provides a systeTn that charges winning bidders only for actual advertising that was del~~~~ed. This mecban, as~ provides a very accurate charging mechanism. However, in practice the estimate systei-n employed in the voting step may accurately estimate audience size, particularly if the ~~-v~.~~ing mechanis~n described below is employed. In this instance the delivery notification sgrstern need not be impl~~~en~ed and the voting estimate may be used in the final price computations, As described above, voting can return a binary match Yes / No match indication.
Some of the targeting mechanisms do have binary resolutions (for instance those based on geography), however other mechanisms (for instance the age aiid gender of the current audience that is detemi1~ed by a classifier system) have probabilistically d~~ennined match criteria. Another voting mechanism is to return the probability (i.e., goodness of fit) that a particular canilaaign matches. The list that is ~~tumed might include a probability for each campaign, or it might return indications for only those campaigns w1~~e the probability exceeds a given tlreshold. Collating the probabilistic votes may be done in a statistical maainer that generates a probability distribution describing the likelihood of the size of an audience for each campaign that was voted for.
Likewise that distribution may be used to calculate an expected value for the revenue that wotild be derived from eac1~ campaign.
As the titne between voting and the actual insertion of advertising increases, so increases the likelihood that the size and cb.srac-ter of the audience has changed. If the difference is only a few minutes (e.go, 2 or 3 minutes), and there hasii't been a prÃ~~am chaiige, tli~ii the difference is probably small. If, on. the other hand, the difference if 15 or 20 minutes it is quite likely that there has been a substantial c1iange. 'rwo altematlves are presented for dealing with the change of audience. The first is to bugld. a probability model of the how an audience changes over time, and use teclriiques such a non-linear filtering to predict t1-ic likely changes in the audience. A second al~ematave is to periodically (for instance every 5 minutes) carry out a revote, and if the result of the new vote is substantially different f-r~~ the previous vote carry out a new auction. Some care needs to be taken to avoid conditions where the actual break happens during the r~-va~~e and ~~-auction process. In suc.h an instance W~~~e a break occurs before a reLLauction is completed, previous auction results may be utilized to identify winning bidders and select assets for insertion.
When multiple simult~~o-us assets are provided to a LiED or UED at one time, the UED must pick one oi'these assets to delivere Alternatives for selecting assets include first match aiid best matcb. In first match mode, asset choices are ordered in the same order in which their respective auctions were won and then the lrl;D selects the first one that is a reasonable match. ln best match mode, the UEDs current estimate for a best match among the altemakives is Ã;b.~~en.
The auction platform is responsible for determining the winner or winners of each auction and the price that each winning bidder should pay. In circumstances where there are multiple winners it may be desirable to incrementally d~~ennine wiiin~s and then d~~~mii~~ the price that they pay after all winners have been determined.
The auctions described in relation to specific avails take place over a period of ti~~~ and allow a bidder to change a bid during the course of the auction..
Tbis is because the goods being sold (the avails) can be detes-inined ahead of time. However, in the case of auctions rLm in a~~~gatis~n mode, this may not ~e possible because the n~niber of ~ealn 24 time viewers is a critical component in the desc.-ription. oi't.h~
audience, and that number is not known until a very short period of time before the asset is distributed.
Complicating matters fi,arther, when multiple positions are being auctioned, the number of viewers for a.
given position may be highly dependent on viewers f.or the otI~e.r positions.
Consider the -followin~ exaniple:
Positive votes are indicated with an X:

Asset I Asset 2 Asset 3 Asset 4 x UED A x UED B x x I;TE~ ~ ~ x -- - --------- -If the bidder owriing Asset l wins the auction then Asset 3 is reduced to only one vote and Asset 4 has no votes. If on the otber 1~and the bidder owning .A,S~~~
3 wins the auction then Asset 2 is reduced to no votes and Asset I is reduced to one vote. 'I'he important observation is that the auction for the second position (e.g., parallel distribution) changes quite drwnatzcally. The consequence of this observation is that the auction. typically runs entirely in an automated mode, ~id that the bidders may not have an opportunity to c1iange their bids during the bidding process (although they may be able to change there bids up to the moment that the auctioii is conducted).
l~ the descriptions of voting models the following two vote tabulations will be used (a I is used to indicted a positive vote). These votes represent 5% of the UED
population:

SeeÃ~ad0 1 Votes for Assets A B c D E

u 7 1 1 15 "1-he following vote tabulation supposes the winning bid specified asset C
and therefore all votes associated witli, asset Cl are removed and new total computed.

Scenario 1 Votes for Assets AFTER C is removed A B C O E
2 Ã ~

--------------------- ---- ----u 7 D 9 1 1 ~

12 1 1 Ã 1 1~
~'O~'~,~ 3 1 0 3 37~

In the following descriptions the term legal bid or legal offering is used to describe a bid that does not violate a bidder's complete bid, which includes the total 5 amount they are willing to pay and any coiis~raints that they have made. For instance, if a bidder has said the maximum that they are willing to pay for an ad campaign is $1,000 and they have already accumulated $990 in advertising, then any subsequent bid of less than or equal to $10 is legal, but any larger bid is not. One novel consequence of this auction model is that all c~~paigiis compete for every avail, and in particular, multiple 10 campaigns for the same bidder may end up bidding against each other.
Special rules may be implemented to prevent this f- s~i-ri bap~eninge In particular, once a particular buyer wins a bid, then for the ctirrer~t a-Lict1on other bids froni that buyer could be coaisidered illegal.

All of these auction models are described assuming that there is only one parallel coiitent difitTibution available in a given avail (i.e, a flotilla with only one column).
floweverj it will be appreciated that multiple parallel distribution may be available.
Ea ~ bldd.er wins For eaeb asset an offerin~prace is calc~alateÃ1 as follows: the maximum bid for a~.
asset is inultlplied with the estimated audience size. The largest legal offering pnce wins the auction. In the case of a tie, one of the bidders may be picked at random or another tieybreaklng mechanism may be impleTnented. The price per impression paid is the maxiflnw-n bid price.

'I'he following exainple takes the total from the scenario 1votes and supposes a range of bids (on a cost per impression (CPI) basis) and determines the winner. Note that since the assuanption in this exainple is that 5% of the set tops vote the estimated audience is 20 times this total vote for eac1i asset. The bidder associated. with asset C wins and pays a C113I of $0.30:
______ _______________________________________________________________________________ ______________~ _s ~ ______ _ ______________..,..-._.-.................................
~t A B C D E
Totaà Vote 7 8 11 6 10 Estimated Audience 140 160 220 120 200 CPI BID 0.30 0.25 0.10 0õ25 .............. .........
C3ffehng Pdce $ 42.00 $ 40.50:::$ 12.Ã30 50.00 An alt~iative set of CPIs can yield a different winner, the bidder associated wit:li asset D who will pay a CPI of $0.60~
Asset l otal Vote 7 8 1f 6 10 Estimated Audience 143 160 220 120 200 ..........
CPÃ BÃD 0.30 0.25 t3e25 Offedne~ Pdce $ 42.Ã30 $ 40.00 $ 66.00 $ 50.00 W I fVNER
This auction is repeated for each parallel distribution opportunity and there is no aqjustmeiit in price.
After C wins in the first example in this section, the votes are recounted and the following detennines the second wi~iier, which in this case is the Asset A
owner who will pay a CPI of $0.30 -----------------------~-.-.-.-....-...~
Asset A B ~ D E
Total VOte 3 1 0 3 3 Estimated Audience 60 20 0 60 60 ~~'I ~~C~ 0.25 0.30 t3. ti~ O,25 Ã~
C4ffer%~ig F~dc~ 5.~30 s - $ 6.it~4 $ 1a.t3~
I 5 6~4~I~JÃ~~~
------------------------HIph bidder wins, VIcIC~~~ pricing For each asset an offering price is calculated as follows: the maximum bid associated with the asset is multiplied with the estimated audience size.
'I'he largest legal offering price wins the auction, and, in the case of a tie, one of the bidders may be picked at random or another basis, or the avail may be split. The estimated total price that they will pay is one laermy more than the next highest legal bid. The winning price per ~~~~~~~~~on is calculated by dividing the estimate total price by the estimated size of the winning bids audience.
Using the votes for scenario I as an example, the winner is again the owner of Asset C. However the amount they will pay is $50 / 220 = $0.227 CP1.
Asset A B c D E
TotalVote 7 8 11 6 10 EstEmated AudEence 140 160 220 920 200 CPI B10 OM 0,25 Ã5.30: 0.10 025 - .~--Cffedng F~~~~ $ 4 2.0 0 $ 4 0uC3 :::.:<;::::~,~ $ 12.0 0 $ 50.00 WINNER
~~s auction is repeated for each parallel distribution o~por-&,-u~~ty and there inay be no adjustflnent in lsiice.

fla h b1dder.win~ ~t ~ same total ri~~

For each asset an offering pnce is calculated as follows: the maximum bid associated with the asset is multiplied with the estimated audience size. ''he largest legal offering pn~e wins the auction. Final price calculation may be done after all winners for a given flotilla are decided.
1 5 The auction is repeated for each parallel distribution o~~~~iilty. Once all winners have been determined then the price paid by the lowest winning bidder is used as the estimated price. 'T'1ie winiiiflig price per impression for each bidder is calc-ulated separately for each as lsv dividing the estimated price by the estimated size of each Winiiing bids audience.
Applying this method to flie votes of scenario 1, and ~~sinnan~ a parallel distribution opportunity for 2 sirnultaneous assets, t1~~~i the winners will be the owners of Assets A and C. Each will pay the price equivalent to $18. 'I'he owiie$ of A
will pay $18 ~ 220:__: $0,081 8 CPI and the owner of A will pay wliat they bid $0,30.
Asset A B c D E
TÃst~l Vote 7 8 11 6 10 EstÃmat~d Audience 140 160 220 120 200 CPÃ BID 0.30 0.25 0.30 0.10 0.25 i~edng Pdce $ 42,00, $ 40,~3~ ;~1 $ 12.0 0 $ 50041 WINNER

Aiid the second auction after C is removed.

... .. . . .... . .......
__ ...
Asset A B c D
'~" E
otal Vote 3 1 0 ~ 3 Estimated Audience 60 20 0 ~Q 60 CPI ~310 .~ 0,25 Ã~.3~} .1~} 0,25 ~~ia~g Pa~ce iiS:::::::1$ 5,0&~ $ ~}
t3 $ 15.00 W INhBER
------------------------- ----------------------High bidd.~rwins5 AR p~~ same price per lMressi0n For each asset an offeriai,~ price is calculated as follows: the maxi~iuni 'bid associated with the asset is multiplied with the estimated audience size.
I'lie largest legal offering pn*ce wins the auction, in the case of a tie, one of the bidders is picked at random.
1~iiial price calculatioa~ may be done after all winners for a given flotilla are decided. 'I'he auction is repeated for each parallel distribution opportunity. Once all winners have been det.emiined then the lowest price paid per impression by a winning bidder is the winning price per impressiort, for each bidder.
Applying this to vote scenario 1, and assarning a parallel distribution opportunity for 2 sim-Laltan~ous assets, then the u imiers will be the owners of Assets A
aiid C. Each will pay the (;PM associated with the lowest winner, which in this case is $0.30.
Asset A B C D E
Total Vote 7 8 11 6 10 Estimated Audience 140 160 220 120 200 CPI BID 0.30 0.25 0.3Ã3 0.10 0.25 ~r~~s~ ~~i~~ $ ~~ eÃ~i3 12.0~3 $ 50.00 ~~~ ~~~ >
l. 5 And the second auctiozi after CI is removed.
Asset A B C D E
Total gdole 3 1 0 3 3 Estimated Audience 60 20 Ã5 60 60 C-1I BI[3 ::: 0.25 Ã1.30 0.10 0.25 Offeris~~ Pdce ~~~~ ~ $ ~S,4~d3 $ $ 6.Ã1~3 $ 15.00 ~!'~'INf~~~
The preceding auction discussions assume only one parallel distribution altemati~e within an avail (break). In general, there will be more than on, e.
A separate auction should be run for each flotilla column, although it should be noted that the pool of votes may need to be updated for the subsequent breaks after an asset is placed (minimum separation rules will usually prevent the same asset from being delivered twice in a row).

Commodity code rules may also make some assets "illegal" after another ad bas been place. One way to run an auction is to sell the contents of each column in a sequential fashion. However, an altemative mechanism is to sequentially auction all of the first positions in eac1i ~lLminy then auction the second positions proceeding in this fas~~~~~~
until all positions have beert, sold.
Considerable ffistor~ca~ ~nfcamiation about auctions accumulates quickly.
'f'his infonnatgon can be used to assist a bidder in making their bids. For instance, historical anfararx.ation about all previous campaigns that match the targeting of a newly created campaign can be retrieved. This ar~f-branatiÃ~~ can suggest the average number of impressions that are available for a giveii type of cairt.paf gn oai a daily basis (as well as the total number of impressions that are available on a daily basis). Average cost per ampressiorf, for similar canipaigns can also be retrieved. Aggregate information about current carnpaigns can also be retrieved and the dernand for impressions can be calculated. This d.eanand can be compared with. the historical dernand and prices to produce a rough estimate of what current prices are likely to be.
When a bidder is entering a new campaign they may req-uest (e.g., via an interface) the system to provide historical ~nforrnat~on and/or for estimates of prices and available impressions. This information could then guide the bidder in the number of impressions that they are likely able to get over a given ~~~e penod a~.d suggest a bidding range that would likely get tha, that amount impressions. Of course, the sys~ern can only provide estimates since external force may increase demand unexpectedly, or supply may reduce or any number of factors may invalidate the estimate. For this reason it may be important that buyers be able to update their bidding parameters as their campaigns progress.
Ca~Raign Monit0~in~
Wbile a particular campaign is active for a bieitter several pieces of info~iatioit can be made available to theirn.
Exairaples of available informatiorg, include:
- Cumulative count of impressions for the campaign - Daily, weekly and monffily iirnpression counts for the carnpaign since it started and, if appropriate, a comparison to goals a.ss~ciated i&ritb. pacing budget - Current status of the budget, botli spent and remaining ftmds~ and similar status for pacing budgets - Daily, Weel~y and monthly total costs for the campaign since it started and, if appropriate, a comparison to pacing budgets.
= Detailed infon-nation abcr-ut all auctions won.
~ Detailed i-n~~rination about auctions t1lat were lost, provide some information about the winning bids (estimates audience sizes and lmpressi~ii costs).
$ Average nuTnber of total 1nipreSsions delivered by the ~ystera per day, week and month.
~~etailed day~~~-day, weekLLby4Week and month-byymonth total impressions delivered by the ss~~tein.
1 pAver.a,~e nu-mber of total impressions delivered by the system per day, week and month for commonly purchased targets. For instance, the most commonly bought age aa~(i gender targets or most coraunonl~ purchased ~~grapbic areas.
-Detail~ day-by-clay, wLvkTb;r-W~k and month-by-month toW impressions delivered by the systern for commonly purdbased targets.
The gn~~rination provided to bidders can be delivered in. a number of different formats. Some of these formats may be more appropriate for some kinds of data.
¾Tabular fornat g~~~~~~heets sGraplis There are numerous dif-T~ent ways in which data may be deliveced. to wz~~~~~ng bidders, by the system. Some of these mecbanlsnis ii'icltacl~ users acccsSir~g d~..~.
interactively dia the internet using a web browser. This maiuier of interactive access would allow usLTs to search for sTs~~~c histofical data if it is useful to ~iem, I)T~~ ~~i also receive periodic email.
messages that sumina.ri,~e the status, of their canipaign. One manner in which these rcpor~ ~~
be made available is to provide a. menu of standard -repÃ~rt types that a user can request be mailed to t1ietn. Of course an option that provides for fLilly customized reports can also be supporte1. Users can also request that periodic fax summaries by sent to ~~in.
Further, users can reqtiest that periodic paper reports be mailed to them. Some buyers may be r-om~~~~
with several different campaigns at once. Additional summary i~fortnat~on that presents the overall status of all, or various subsets, of their active campaigais can be Swnmarized and made available to them.

Wliil~ various embodiments of the present invention have been described in detail, 1 arther modifications atid adaptations of the invention may occur to those skilled in the art. However, it is to be expressly understood that such modifications and adaptations are within the spirit and scope of the present invention.
`1'he foregoing description of the present invention has been presented for purposes of illustration and description. ~~~en-nore9 the description is not intended to limit the invention to the form disclosed herein. Consequently, variations and modifications co~nmensu~~~~ ~it1i the above teachings, and skill and knowledge of the relevant art, are within the scope of the preseait invention. The embodiments described hereinabove are ~'~.rkher intended to explain best modes known of practicing the invention and to enable others skilled in the art to Liti1ize the invention in such, or other eTnlsodainents ~id with various modifications ~~q-uiresl by the particular application(s) or use(s) of the present in~entiono It is intended that the appended claims be construed to include a1tema1.~~e embodiments to the extent permitted by the pi%or arL

Claims (69)

1. A method for use in connection with delivering assets to users of a broadcast network, the broadcast network primarily involving synchronized distribution of broadcast content to multiple users, the method comprising:
providing information regarding an available asset delivery spot for delivering content;
receiving bids for said asset delivery spot, wherein said bids are received from one or more asset providers;
based on a winning bid, inserting an asset of one of the asset providers into the a content stream of the broadcast network for delivery during said asset delivery spot; and adjusting a payment associated with the winning bid based on information regarding deliveries of said asset to users of the broadcast network.

2. The method of Claim 1, wherein providing comprises providing network channel information and delivery time information for said asset delivery spot.

3. The method of Claim 1, wherein providing further comprises:
providing ratings information for said asset delivery spot.

4. The method of Claim 2, wherein said ratings information is provided for at least a first characteristic of a network users receiving the available asset delivery spot.

5. The method of Claim 1, wherein providing comprises providing ratings information for at least first and second different demographic groups of network users.

6. The method of Claim 1, wherein providing comprises providing at least one audience classification parameter for said asset delivery spot.

7. The method of Claim 6, wherein said audience classification parameter relates to at least one of age, gender, income, geographic locale of a portion of an audience segment of interest.

8. The method of Claim 1, wherein receiving said bids comprises interactively receiving bids from one or more user interfaces associated with one or more asset providers.

9. The method of Claim 1, wherein receiving bids further comprises:
receiving predefined ad campaigns from one or more asset providers, said ad campaigns having one or more targeting criteria and bid information; and correlating said targeting criteria and said bid information of said predefined ad campaigns with the the information regarding an available asset delivery spot to identify a winning bid.

10. The method of Claim 9, wherein correlating is an automated process.

11. The method of Claim 10, wherein said winning bid is determined during a program that includes the asset delivery spot.

12. The method of Claim 1, wherein inserting comprises:
replacing a default asset in the content delivery stream with the asset associated with the winning bid.

13. The method of Claim 1, wherein, inserting comprises:
transmitting instructions to at least a portion of UEDs within the broadcast network to play the asset associated with the winning bid during said asset delivery spot, wherein said asset has previously been stored in storage associated with said UEDs.

14. The method of Claim 1, wherein inserting comprises:
inserting the asset associated with the winning bid into a parallel content stream that is broadcast in synchrony with a content stream including the asset delivery spot.

15. The method of Claim 14, further comprising:
providing information to at least a portion of UEDs in said broadcast network regarding the availability of said asset on said parallel content stream, wherein the information includes at least one target criteria for said asset.

16. The method of Claim 1, further comprising:
based on a second winning bid, inserting a second asset of one of the asset providers into the a second content stream of the broadcast network for delivery during said asset delivery spot.

17. The method of Claim 1, further comprising:
receiving asset delivery notifications from at least a portion of UEDs in the broadcast network, said asset delivery notifications providing indication of actual deliveries of said asset to network users.

18. The method of Claim 17, wherein said winning bid is adjusted based on said actual deliveries.

19. The method of Claim 1, wherein said winning bid comprises the highest bid for said asset delivery spot.

20. The method of Claim 19, wherein said winning bid is adjusted to a nominal amount more than the second highest bid.

21. A method for use in connection with delivering assets to users of a broadcast network, the broadcast network primarily involving synchronized distribution of broadcast content to multiple users, the method comprising:
auctioning an asset delivery spot of a broadcast program for broadcast to a network audience, wherein said auctioning is based on at least first and second audience characteristics;
inserting first and second assets associated with winning bids for said first and second audience characteristics into at least one content stream of the broadcast network;
and delivering, via said broadcast network, said first asset to a first portion of the network audience of said broadcast program during said asset delivery spot;
delivering, via said broadcast network, said second asset to a second portion of the network audience of said broadcast program during said asset delivery spot.

22. The method of Claim 21, wherein said first and second assets are delivered simultaneously.

23. The method of Claim 21, wherein said first and second assets are delivered on separate broadcast content streams.

24. The method of Claim 21, wherein auctioning comprises:
performing a first auction for said first audience characteristic; and performing a second auction for said second audience characteristic.

25. The method of Claim 21, wherein auctioning comprises:
providing information regarding said asset delivery spot; and receiving bids for said asset delivery spot, wherein said bids are received from one or more asset providers.

26. The method of Claim 25, wherein providing information further comprises:
providing ratings information for said asset delivery spot.

27. The method of Claim 26, wherein said ratings information is provided for at least first and second different demographic groups of network audience.

28. The method of Claim 21, wherein said audience characteristics relate to at least one of age, gender, income, geographic locale of a portion of an audience segment of interest.

29. The method of Claim 21, wherein inserting comprises:

replacing a default asset in the content delivery stream with one of the first and second assets and inserting the other of the first and second assets into a parallel content stream.

30. The method of Claim 29, wherein, inserting comprises:
transmitting instructions to at least a portion of UEDs within the broadcast network to play one of the first and second assets during said asset delivery spot, wherein said first and second assets have previously been stored in storage associated with said UEDs.

31. The method of Claim 31, wherein inserting comprises:
inserting the first and second assets into first and second parallel content streams that are broadcast in synchrony with a content stream including the asset delivery spot.

32. The method of Claim 31, further comprising:
providing information to at least a portion of UEDs in said broadcast network regarding the availability of said first and second assets on said parallel content streams, wherein the information includes at least one target criteria for each said asset.

33. The method of Claim 32, wherein delivering said first and second assets comprises:
allowing UEDs of said broadcast network to select one of the first and second assets for delivery during said asset delivery spot based on said target criteria.

34. The method of Claim 21, wherein delivering comprises:
allowing UEDs of said broadcast network to randomly select one of the first and second assets for delivery during said asset delivery spot.

35. The method of Claim 21, further comprising:
receiving asset delivery notifications from at least a portion of UEDs in the broadcast network, said asset delivery notifications providing indication of actual deliveries of said first and second assets to said first and second portions of said network audience.

36. The method of Claim 21, further comprising:
adjusting payments associated with the winning bids based on information regarding deliveries of said first and second assets to said network audience.

37. The method of Claim 21, wherein auctioning further comprises:
receiving predefined ad campaigns from asset providers, said ad campaigns having one or more targeting criteria and bid informations and correlating said targeting criteria and said bid information of said predefined ad campaigns with information regarding said asset delivery spot to identify said winning bids.

38. The method of Claim 37, wherein correlating is an automated process.

39. The method of Claim 37, wherein said winning bids are determined during a program that includes the asset delivery spot.

40. A method for use in connection with delivering assets to users of a broadcast network, the broadcast network primarily involving synchronized distribution of broadcast content to multiple users, the method comprising:
offering a predetermined number of audience impressions for an audience having at least one predefined targeting criteria;
receiving bids for said audience impressions from one or more asset providers;

and delivering an asset of a winning asset provider to network users having said at least one defined targeting criteria.

41. The method of Claim 40, wherein said asset is delivered until said predetermined number of said audience impressions is achieved.

42. The method of Claim 41, wherein delivering comprises:
delivering said asset at two or more temporally distinct times.

43. The method of Claim 41, wherein delivering comprises:
delivering said asset on two or more different network channels.

44. The method of Claim 40, wherein delivering comprises:
delivering said asset to said predetermined number of said network users within a predefined time frame.

45. The method of Claim 40, wherein said targeting criteria relates to at least one of age, gender, income, geographic locale of a portion of an audience segment of interest.

46. The method of Claim 40, wherein receiving said bids comprises interactively receiving bids from one or more user interfaces associated with one or more asset providers.

47. The method of Claim 40, wherein receiving bids further comprises:
receiving predefined ad campaigns from one or more asset providers, said ad campaigns having one or more targeting criteria and bid information; and correlating said targeting criteria of said predefined ad campaigns with the targeting criteria for said predetermined number of audience impressions to identify a winning bid of said winning asset provider.

48. A method for use in connection with delivering assets to users of a broadcast network, the broadcast network primarily involving synchronized distribution of broadcast content to multiple users, the method comprising:
obtaining at least one audience characteristic for an upcoming asset delivery spot in a broadcast content stream;
identifying ad campaigns having targeting criteria corresponding with said at least one audience characteristic;
determining a winning ad campaign from said identified ad campaigns; and delivering an asset associated with said winning ad campaign during said asset delivery spot.

49. The method of Claim 48, further comprising:
adjusting at least one parameter of said winning ad campaign based on said delivering of said assert.

50. The method of Claim 49, wherein said at least one parameter relates to a number of delivered impressions for said winning ad campaign.

51. The method of Claim 49, wherein said at least one parameter relates to costs for said winning ad campaign.

52. The method of Claim 48, further comprising:
obtaining information regarding a number of network users who received said asset.

53. The method of Claim 52, further comprising calculating a cost for delivering said asset during said asset delivery spot based at least in part on said number of network users who received said asset.

54. The method of Claim 53, wherein calculating said cost comprises multiplying said number of network users -who received the asset with a bid price associated with said winning ad campaign

55. The method of Claim 52, wherein obtaining information comprises:
obtaining statistical information associated with delivery of said asset to said network users.

56. The method of Claim 52, wherein obtaining information comprises:
obtaining reports from UEDs of network users who received said asset.

57. The method of Claim 48, wherein obtaining at least one audience characteristic includes obtaining information relating to audience size.

58. The method of Claim 48, wherein obtaining at least one audience characteristic includes obtaining information relating to audience demographics.

59. The method of Claim 48, wherein identifying further comprises identifying ad campaigns having temporal targeting criteria corresponding with a time associated with said asset delivery spot.

60. The method of Claim 48, wherein identifying further comprises identifying ad campaigns having programming targeting criteria corresponding with programming associated with said asset delivery spot.

61. The method of Claim 48, wherein determining the winning ad campaign comprises:

multiplying a bid price associated with each ad campaign with an estimated audience for said asset delivery spot to determine an offering price for each ad campaign;
and selecting at least one winning ad campaign based on said offering price.

62. The method of Claim 61, wherein said estimated audience is separated into different estimated audiences by demographic group and different ad campaigns may be multiplied by said different estimated audiences to determine said offering price for each ad campaign.

63. The method of Claim 48, wherein said winning ad campaign is determined during a program that includes the asset delivery spot.

64. A system that delivers assets to users of a broadcast network, the broadcast network primarily involving synchronized distribution of broadcast content to multiple users, the system comprising:

a network interface for receiving ad campaign information from asset providers;
an auctioning processor that processes said ad campaign information to identify a winning ad campaign for at least a first asset delivery spot;
a content synchronizer that, when directed by the auctioning processor, delivers an asset associated with the winning ad campaign to users of the broadcast network; and a traffic and billing module that calculates a value for the delivery of the asset based on the delivery of the asset to the users of the broadcast network.

65. The system of Claim 64, wherein delivery of the asset includes an insertion of the asset into a stream of programming associated with said asset delivery spot.

66. The system of Claim 64, wherein delivery of the asset includes an insertion of said asset into a content stream that is broadcast in synchrony with a content stream including programming containing said asset delivery spot.

67. The system of Claim 64, wherein said auctioning processor is operative to interactively receive ad campaign information form said asset providers during an auction.

68. The system of Claim 64, wherein said auctioning processor is operative to utilize predetermined ad campaign information to determine one or more winning ad campaigns for said asset delivery spot.

69. The system of Claim 64, wherein said traffic and billing module is operative to receive asset delivery reports from customer premise equipment of users of the broadcast network.