METHODS AND SYSTEMS FOR AD PLACEMENT PLANNING

- Yahoo

Methods and systems are disclosed which allow shifting inventory to fulfill guaranteed delivery advertisement contracts. Inventory may be allocated from a supply of unallocated inventory to one or more advertisers in accordance with guaranteed delivery agreements. Inventory may be reserved for the one or more advertisers from the remaining supply of unallocated inventory. Inventory may then be allocated to an additional advertiser by using unallocated inventory or shifted inventory, or a combination of unallocated inventory and shifted inventory. The shifted inventory is shifted out of the allocation for the first advertiser and the shifted inventory is replaced by the reserve inventory for the respective advertiser.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
CROSS REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. application Ser. No. 13/339,254 filed Dec. 28, 2011, the entire content of which is hereby incorporated herein by reference.

BACKGROUND

Online advertising marketplaces and exchanges may involve, for example, entities or parties including advertisers, publishers and data providers, as well as a marketplace or online advertising operations facilitator, or market-maker. Data providers may supply information, such as information regarding users, user behavior or user interests, which may enhance value to advertisers in connection with purchasing of advertising inventory. However, in marketplaces and exchanges, significant difficulty and transactional friction may exist in selectivity, arrangements and cooperation between parties. This can lead to suboptimal interactions, reducing efficiency and disincentivizing maximum engagement and spending. Furthermore, friction as well as inequities or unfairness may exist in connection with pricing arrangement, actual pricing, and allocation of spend between parties including advertisers and data providers, and such as in connection with sold advertising inventory.

There is a need for improved techniques in online advertising, including in online advertising marketplaces and exchanges, to increase transactional efficiency.

SUMMARY

Some embodiments of the invention provide systems and methods for inventory reservation to fulfill guaranteed delivery advertisement contracts. In accordance with some embodiments, a first inventory may be allocated from a supply of unallocated inventory to a first advertiser, wherein the first inventory satisfies the first advertiser's targeting criteria, and wherein the supply includes advertising opportunity inventory. For example, an advertiser may request 5000 impressions from a supply of 10,000 impressions. In some embodiments, the relationship between demand (e.g., contracts) and supply (e.g., impressions) may be modeled as a bipartite graph. A second inventory is then allocated from the remaining supply of unallocated inventory to a second advertiser, wherein the second inventory satisfies the second advertiser's targeting criteria.

A third inventory is reserved from the remaining supply of unallocated inventory to the first advertiser. The third inventory may be reserved for the first advertiser proportionally. In other words, the amount of the reserve inventory allocated to the first advertiser may be proportional to the amount of percentage of matching inventory that was allocated to the first advertiser. For example, out of a 10,000 impression supply, if 5000 was allocated to the first advertiser, and all of the remaining 5000 impressions meet the first advertiser's targeting criteria, then 2500 impressions (50% of 5000) will be reserved for the first advertiser. A fourth inventory is then reserved from the remaining supply of unallocated inventory to the second advertiser, wherein the fourth inventory satisfies the second advertiser's targeting criteria. As explained above, the fourth inventory may be reserved for the second advertiser proportionally.

A fifth inventory is allocated to a third advertiser, wherein the fifth inventory satisfies the third advertiser's targeting criteria and also satisfies the first advertiser's targeting criteria or the second advertiser's targeting criteria, by using unallocated inventory or shifted inventory, or a combination of unallocated inventory and shifted inventory. The shifted inventory is shifted out of the allocation for the first advertiser or the allocation for the second advertiser, and wherein the shifted inventory is replaced by the reserve inventory for the respective advertiser. Using the above example, assume that 5000 impressions from the supply of 10,000 impressions are allocated to the first advertiser, and 3000 impressions from the remaining 5000 impressions are allocated to the second advertiser. Now, if the third advertiser requests 2000 impressions, and the remaining 2000 impressions meet the third advertiser's criteria, then those 2000 impressions will be allocated to the third advertiser. However, if the 2000 impressions don't meet the third advertiser's criteria, then the third advertiser's demand may be filled with inventory which is shifted from the impressions allocated to the first and/or second advertisers. Thus, if part of the impressions which are allocated to the first and/or second advertisers meet the third advertiser's criteria, those impressions will be allocated to the third advertiser and the impressions which were shifted out from the first and/or second advertisers will be replaced by the impressions which were reserved for the first and/or second inventory.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a distributed computer system according to one embodiment of the invention;

FIG. 2 is a flow diagram illustrating a method according to one embodiment of the invention;

FIG. 3 is a flow diagram illustrating a method according to one embodiment of the invention;

FIG. 4 is a flow diagram illustrating a method according to one embodiment of the invention; and

FIG. 5 is an exemplary bipartite graph according to one embodiment of the invention.

DETAILED DESCRIPTION

FIG. 1 is a distributed computer system 100 according to one embodiment of the invention. The system 100 includes user computers 104, advertiser computers 106 and server computers 108, all coupled or able to be coupled to the Internet 102. Although the Internet 102 is depicted, the invention contemplates other embodiments in which the Internet is not included, as well as embodiments in which other networks are included in addition to the Internet, including one more wireless networks, WANs, LANs, telephone, cell phone, or other data networks, etc. The invention further contemplates embodiments in which user computers 104 may be or include desktop or laptop PCs, as well as, wireless, mobile, or handheld devices such as cell phones, PDAs, tablets, etc.

Each of the one or more computers 104, 106 and 108 may be distributed, and can include various hardware, software, applications, algorithms, programs and tools. Depicted computers may also include a hard drive, monitor, keyboard, pointing or selecting device, etc. The computers may operate using an operating system such as Windows by Microsoft, etc. Each computer may include a central processing unit (CPU), data storage device, and various amounts of memory including RAM and ROM. Depicted computers may also include various programming, applications, algorithms and software to enable searching, search results, and advertising, such as graphical or banner advertising as well as keyword searching and advertising in a sponsored search context. Many types of advertisements are contemplated, including textual advertisements, rich advertisements, video advertisements, etc.

As depicted, each of the server computers 108 includes one or more CPUs 110 and a data storage device 112. The data storage device 112 includes a database 116 and an Ad Placement Planning Program 114.

The Program 114 is intended to broadly include all programming, applications, algorithms, software and other and tools necessary to implement or facilitate methods and systems according to embodiments of the invention. The elements of the Program 114 may exist on a single server computer or be distributed among multiple computers or devices.

As will be understood by one of ordinary skill in the art, in a guaranteed delivery advertising model, advertisers may enter into an agreement (e.g., a contract) with a publisher (e.g., a search engine, online newspaper, etc.) which guarantees or provides some measure of assurance that the advertiser will receive a certain agreed upon amount of suitable advertising. For example, the agreement may provide that the advertisement will have one million impressions (e.g., page views). By contrast, non-guaranteed delivery advertising models may be based on individual serving opportunities or may be spot market-based. In various models, advertisers may pay based on any of various metrics associated with advertisement delivery or performance, or associated with measurement or approximation of a particular advertiser goal. For example, models can include, among other things, payment based on cost per impression or number of impressions, cost per click or number of clicks, cost per action for some specified action, cost per conversion or purchase, or cost based on some combination of metrics, which can include online or offline metrics. The guaranteed delivery contracts may define targeting criteria such as, for example, user location, age, demographic, user profile information, search history, browse history, etc. For example, an advertiser may request that their advertisement be displayed when a page is visited by young females in CA.

However, the “supply” of impressions is substantially uncertain and may change over time. For example, unpredictable events such as natural disasters or acts of terrorism may result in large increases in traffic. Thus, since the supply is based on the impressions, the publisher can only attempt to estimate the future supply. In addition, the capacity is non-storable over time; as the supply is generated from the number of impressions, the publisher has to decide how to allocate the supply within a few seconds before it perishes. Selling the impressions to potential advertisers is further complicated by the issue of contention between different advertisers for the same impression. Advertisers may seek to reach a specific audience, e.g., females from California before sunset and when the stock market is up (e.g., Dow Jones Industrial Average has risen). Others may accept more loosely defined, less targeted impressions (e.g., females).

Embodiments of the invention improve the efficiency of booking and allocating maximum demand and checking for availability for new guaranteed delivery contracts while managing contention between different contracts. FIG. 2 is a flow diagram illustrating a method 200 according to one embodiment of the invention. At step 202, using one or more computers, a first inventory from a supply of unallocated inventory is allocated to a first advertiser, wherein the first inventory satisfies the first advertiser's targeting criteria, and wherein the supply includes advertising opportunity inventory. For example, an advertiser may request 5000 impressions from a supply of 10,000 impressions. In some embodiments, the relationship between demand (e.g., contracts) and supply (e.g., impressions) may be modeled as a bipartite graph (explained in more detail in relation with FIG. 5).

At step 204, using one or more computers, a second inventory is allocated from the remaining supply of unallocated inventory to a second advertiser, wherein the second inventory satisfies the second advertiser's targeting criteria.

At step 206, using one or more computers, a third inventory is reserved from the remaining supply of unallocated inventory to the first advertiser, wherein the third inventory satisfies the first advertiser's targeting criteria. The third inventory may be reserved for the first advertiser proportionally. In other words, the amount of the reserve inventory allocated to the first advertiser may be proportional to the amount of inventory that was allocated to the first advertiser in step 202. For example, out of a 10,000 impression supply, if 5000 was allocated to the first advertiser, and all of the remaining 5000 impressions meet the first advertiser's targeting criteria, then 2500 impressions (50% of 5000) will be reserved for the first advertiser.

At step 208, using one or more computers, a fourth inventory is reserved from the remaining supply of unallocated inventory to the second advertiser, wherein the fourth inventory satisfies the second advertiser's targeting criteria. As explained above, the fourth inventory may be reserved for the second advertiser proportionally.

At step 210, using one or more computers, a fifth inventory is allocated to a third advertiser, wherein the fifth inventory satisfies the third advertiser's targeting criteria and also satisfies the first advertiser's targeting criteria or the second advertiser's targeting criteria, by using unallocated inventory or shifted inventory, or a combination of unallocated inventory and shifted inventory, wherein the shifted inventory is shifted out of the allocation for the first advertiser or the allocation for the second advertiser, and wherein the shifted inventory is replaced by the reserve inventory for the respective advertiser. Using the above example, assume that 5000 impressions from the supply of 10,000 impressions are allocated to the first advertiser, and 3000 impressions from the remaining 5000 impressions are allocated to the second advertiser. Now, if the third advertiser requests 2000 impressions, and the remaining 2000 impressions meet the third advertiser's criteria, then those 2000 impressions will be allocated to the third advertiser. However, if the 2000 impressions don't meet the third advertiser's criteria, then the third advertiser's demand may be filled with inventory which is shifted from the impressions allocated to the first and/or second advertisers. Thus, if part of the impressions which are allocated to the first and/or second advertisers meet the third advertiser's criteria, those impressions will be allocated to the third advertiser and the impressions which were shifted out from the first and/or second advertisers will be replaced by the impressions which were reserved for the first and/or second inventory. It should be noted that the above steps 206-210 may be repeated each time a new contract is to be filled (i.e., have inventory allocated to it).

FIG. 3 is a flow diagram illustrating a method 300 according to one embodiment of the invention. At step 302, using one or more computers, a first inventory from a supply of unallocated inventory is allocated to a first advertiser, wherein the first inventory satisfies the first advertiser's targeting criteria, and wherein the supply includes advertising opportunity inventory. For example, an advertiser may request 5000 impressions from a supply of 10,000 impressions. In some embodiments, the relationship between demand (e.g., contracts) and supply (e.g., impressions) may be modeled as a bipartite graph (explained in more detail in relation with FIG. 5).

At step 304, using one or more computers, a second inventory is allocated from the remaining supply of unallocated inventory to a second advertiser, wherein the second inventory satisfies the second advertiser's targeting criteria.

At step 306, using one or more computers, a third inventory is reserved from the remaining supply of unallocated inventory to the first advertiser, wherein the third inventory satisfies the first advertiser's targeting criteria.

At step 308, using one or more computers, a fourth inventory is reserved from the remaining supply of unallocated inventory to the second advertiser, wherein the fourth inventory satisfies the second advertiser's targeting criteria.

At step 310, using one or more computers, a fifth inventory is allocated to a third advertiser, wherein the fifth inventory satisfies the third advertiser's targeting criteria and also satisfies the first advertiser's targeting criteria or the second advertiser's targeting criteria, by using unallocated inventory or shifted inventory, or a combination of unallocated inventory and shifted inventory, wherein the shifted inventory is shifted out of the allocation for the first advertiser or the allocation for the second advertiser, and wherein the shifted inventory is replaced by the reserve inventory for the respective advertiser. At step 312, using one or more computers, one or more advertisements may be served to users (e.g., visitors to websites) based at least in part on the allocations. It should be noted that the above steps 306-310 may be repeated each time a new contract is to be filled (i.e., have inventory allocated to it).

FIG. 4 is a flow diagram illustrating a method 400 according to one embodiment of the invention. At step 402, using one or more computers, a first inventory from a supply of unallocated inventory is allocated to a first advertiser, wherein the first inventory satisfies the first advertiser's targeting criteria, and wherein the supply includes advertising opportunity inventory. For example, an advertiser may request 5000 impressions from a supply of 10,000 impressions. In some embodiments, the relationship between demand (e.g., contracts) and supply (e.g., impressions) may be modeled as a bipartite graph (explained in more detail in relation with FIG. 5).

At step 404, using one or more computers, a second inventory is allocated from the remaining supply of unallocated inventory to a second advertiser, wherein the second inventory satisfies the second advertiser's targeting criteria.

At step 406, using one or more computers, a third inventory is reserved from the remaining supply of unallocated inventory to the first advertiser, wherein the third inventory satisfies the first advertiser's targeting criteria. The third inventory is reserved for the first advertiser proportionally. As discussed above, the amount of the reserve inventory allocated to the first advertiser is proportional to the amount of inventory that was allocated to the first advertiser in step 402. For example, out of a 10,000 impression supply, if 5000 was allocated to the first advertiser, and all of the remaining 5000 impressions meet the first advertiser's targeting criteria, then 2500 impressions (50% of 5000) will be reserved for the first advertiser.

At step 408, using one or more computers, a fourth inventory is reserved from the remaining supply of unallocated inventory to the second advertiser, wherein the fourth inventory satisfies the second advertiser's targeting criteria. As explained above, the fourth inventory may be reserved for the second advertiser proportionally.

At step 410, using one or more computers, a fifth inventory is allocated to a third advertiser, wherein the fifth inventory satisfies the third advertiser's targeting criteria and also satisfies the first advertiser's targeting criteria or the second advertiser's targeting criteria, by using unallocated inventory or shifted inventory, or a combination of unallocated inventory and shifted inventory, wherein the shifted inventory is shifted out of the allocation for the first advertiser or the allocation for the second advertiser, and wherein the shifted inventory is replaced by the reserve inventory for the respective advertiser. It should be noted that the above steps 406-410 may be repeated each time a new contract is to be filled (i.e., have inventory allocated to it).

FIG. 5 illustrates an exemplary bipartite graph according to one embodiment of the invention. Bipartite graph 500 shows the eligibility edges between demand and supply nodes. In other words, bipartite graph 500 depicts how much of the available supply is eligible for each demand node (contract). The table below summarizes the demand and eligible supply for the demand and supply nodes shown in FIG. 5.

Node Demand Eligible Supply D1 200 1000 D2 100 1200 D3 400 1400

The bipartite graph may be mathematically represented as G=(C, S, E) with C representing contracts (e.g., demand nodes) and S representing the supply nodes (e.g., different types of users visiting a website). A demand node j is connected (via edges E) to all supply nodes that are eligible for it. Each supply node i has an associated weight Si that is the forecast for the number of impressions of that type and each demand node j has an associated demand dj which is the number of requested impressions. A feasible allocation for the demands is a way of dividing up the supply among the contracts so that all demands are met. Formally, an allocation is represented by a fraction xijε[0, 1] for every edge (i,j)εE, which specifies that sixij amount of supply i is booked for demand node j. The requirement that the supply node is not overallocated corresponds to the condition that for all supply nodes i, Σj:j˜ixij≦1, where j˜i is notation for “j is connected to i” (and vice-versa). The requirement that all demands are met corresponds to the condition that for all demand nodes j, Σi:i˜jsixij=dj. Suppose an existing graph has an allocation meeting the current demands. We now want to add a new contract to the graph. The new demand node may be represented as j*, with its associated edges (i,j*) for the supply nodes i that are eligible for it.

Embodiments of the present invention allow for the possibility of shifting supply to allow more demand to be booked. The main impediment to shifting is that access is only available to a two neighborhood graph Gj*, and any shifted inventory must necessarily be moved outside the Gj*. Embodiments of the present invention rely on reserve inventories to allow shifting. In accordance with one embodiment, unallocated inventory within every supply node is divided up among the contracts (demand nodes) that are connected to it as reserved inventory. For example, suppose that impression i has two eligible contracts, 1 and 2, and that xi1=0.1 and xi2=0.2. Then supply node i has 70% of its inventory unallocated, and this 70% will be divided between contracts 1 and 2 proportionally. This process is repeated for every impression. In more detail, suppose we start with a graph G=(C, S, E) with some initial allocation, xij for every edge (i,j)εE. Reserve inventories for every contract j in the graph are computed as follows. For every supply node i, the unallocated space among the contracts connected to it is divided up proportional to their allocated amounts. For example, contract j gets

x ij j i x ij

fraction of the unallocated space in i reserved to it, which works out to a reserved supply of size:

Res ij = s i x ij j i x ij - s i x ij ( 1 )

For every contract j, the total reserved inventory to it is:


Rj:=Σi:i˜jResij  (2)

As mentioned above, for a query node j*, a two neighborhood graph Gj* (which shows a set of eligible impressions for j*, as well as all the contracts that match at least one of those eligible impressions) is determined. Impressions i˜j* are referred to as being inside the query region (and other impressions as outside). The allocation xij is computed for every impression i and contract j in Gj*. From this it is determined how much of the “inside” inventory is currently booked. In particular, there is Σi:i˜j*si eligible supply, but Σji:i˜j sixij) of it is booked. However, for each contract j, there is Rj reserved inventory available for shifting. That is, up to Rj inventory currently allocated to j may be shifted from “inside” impressions to “outside” impressions. However, some of this reserved inventory actually comes from inside impressions, so it cannot be used for shifting. Thus, the amount of “inside” reserved inventory is calculated for j, and R′j is set to be the amount “outside.” Thus, for each j, if R′j is greater than the amount currently allocated to inside impressions, then all of the allocation may be shifted outside the query region (recall that reserved inventory is partitioned, so that no two contracts share any reserved inventory). Otherwise, if R′j is less than the amount currently allocated inside, R′j may be shifted outside. Thus, the total amount booked, after shifting as much as possible, becomes:


Σj max{0,Σi:i˜jsixij−R′j}  (3)

Now suppose we are required to book some demand D for the contract j*. If D can be accommodated without changing any previous allocation, then D will be accommodated.

On the other hand, if contracts must be shifted to accommodate D, then the total amount that must be shifted to “outside” inventory is computed. This amount is then split among allocated contracts, proportional to the total amount that each contract is able to shift (this is min{Σi:i˜jsixij, R′j}, where R′j is the reserved inventory “outside.”).

Suppose contract j must shift Shj inventory, and it has X allocated inside. Then, the allocation xij is uniformly changed for every inside impression to be 1−Shhj/Xj. Likewise, since Shhj inventory was shifted outside, the value R′j must be decreased accordingly.

While the invention is described with reference to the above drawings, the drawings are intended to be illustrative, and the invention contemplates other embodiments within the spirit of the invention.

Claims

1. A computerized method comprising:

allocating an inventory of advertising impressions from a plurality of supply nodes to satisfy a demand for advertising impressions received from a plurality of demand nodes, wherein each demand node receives an allocation from at least some of the plurality of supply nodes, and wherein the inventory is allocated in accordance with advertiser targeting criteria associated with each demand node, respectively;
determining a reserve inventory of unallocated advertising impressions for each demand node such that, for each supply node, unallocated advertising impressions are partitioned for allocation among the demand nodes receiving an allocation from a supply node proportionally to the supply node's allocated inventory; and
allocating, for each new demand node, an additional inventory of advertising impressions, wherein the additional inventory satisfies advertiser targeting criteria associated with a new demand node and also satisfies advertiser targeting criteria associated with at least one other demand node, by shifting supply to accommodate the new demand node using a combination of unallocated inventory and shifted inventory, wherein the shifted inventory is shifted out of at least one other demand node's allocation, and wherein the shifted inventory is replaced by the reserve inventory partitioned for allocation to the at least one other demand node.

2. The method of claim 1, further comprising modeling inventory allocation as a bipartite graph such that eligibility edges between the plurality of supply nodes and the plurality of demand nodes depict how much inventory is available for each demand node.

3. The method of claim 2, wherein each demand node is connected via an eligibility edge to one or more eligible supply nodes based at least on advertiser targeting criteria associated with each demand node, respectively.

4. The method of claim 2, wherein, for each eligibility edge, an allocation is represented by a fraction which specifies that an amount of inventory is reserved for a demand node.

5. The method of claim 1, wherein each demand node has an associated demand which is a number of requested advertising impressions.

6. The method of claim 5, wherein, the inventory of advertising impressions is divided among the plurality of demand nodes such that all demands for requested advertising impressions are met.

7. The method of claim 1, wherein each supply node includes an associated weight that is a forecast for a number of impressions of a particular type.

8. The method of claim 1, wherein, for a query demand node, a set of eligible impressions inside a query region are considered to be inside impressions, and other impressions outside the query region are considered to be outside impressions, and wherein for each demand node there is at least some reserved inventory, Rj, available for shifting, such that up to Rj inventory currently allocated to a demand node may be shifted from inside impressions to outside impressions.

9. The method of claim 8, wherein, for each demand node, if an amount of outside impressions, R′j, is greater than an amount of currently allocated inside impressions, then all of the allocation is shifted outside the query region, and wherein if R′j is less than the amount of currently allocated inside impressions, then R′j is shifted outside the query region.

10. The method of claim 1, further comprising:

determining an amount of supply that must be shifted to accommodate a demand that cannot be accommodated without changing a previous allocation; and
splitting the amount of supply that must be shifted among allocated inventory proportional to a total amount of allocated inventory that each demand node is able to shift.

11. A system comprising:

one or more databases;
one or more server computers in communication with the one or more databases, the one or more server computers configured to:
allocate an inventory of advertising impressions from a plurality of supply nodes to satisfy a demand for advertising impressions received from a plurality of demand nodes, wherein each demand node receives an allocation from at least some of the plurality of supply nodes, and wherein the inventory is allocated in accordance with advertiser targeting criteria associated with each demand node, respectively;
determine a reserve inventory of unallocated advertising impressions for each demand node such that, for each supply node, unallocated advertising impressions are partitioned for allocation among the demand nodes receiving an allocation from a supply node proportionally to the supply node's allocated inventory; and
allocate, for each new demand node, an additional inventory of advertising impressions, wherein the additional inventory satisfies advertiser targeting criteria associated with a new demand node and also satisfies advertiser targeting criteria associated with at least one other demand node, by shifting supply to accommodate the new demand node using a combination of unallocated inventory and shifted inventory, wherein the shifted inventory is shifted out of at least one other demand node's allocation, and wherein the shifted inventory is replaced by the reserve inventory partitioned for allocation to the at least one other demand node.

12. The system of claim 11, further comprising modeling inventory allocation as a bipartite graph such that eligibility edges between the plurality of supply nodes and the plurality of demand nodes depict how much inventory is available for each demand node.

13. The system of claim 12, wherein each demand node is connected via an eligibility edge to one or more eligible supply nodes based at least on advertiser targeting criteria associated with each demand node, respectively.

14. The system of claim 12, wherein, for each eligibility edge, an allocation is represented by a fraction which specifies that an amount of inventory is reserved for a demand node.

15. The system of claim 11, wherein each demand node has an associated demand which is a number of requested advertising impressions.

16. The system of claim 15, wherein, the inventory of advertising impressions is divided among the plurality of demand nodes such that all demands for requested advertising impressions are met.

17. The system of claim 11, wherein each supply node includes an associated weight that is a forecast for a number of impressions of a particular type.

18. The system of claim 11, wherein, for a query demand node, a set of eligible impressions inside a query region are considered to be inside impressions, and other impressions outside the query region are considered to be outside impressions, and wherein for each demand node there is at least some reserved inventory, Rj, available for shifting, such that up to Rj inventory currently allocated to a demand node may be shifted from inside impressions to outside impressions.

19. The system of claim 18, wherein, for each demand node, if an amount of outside impressions, R′j, is greater than an amount of currently allocated inside impressions, then all of the allocation is shifted outside the query region, and wherein if R′j is less than the amount of currently allocated inside impressions, then R′j is shifted outside the query region.

20. The system of claim 11, further comprising:

determining an amount of supply that must be shifted to accommodate a demand that cannot be accommodated without changing a previous allocation; and
splitting the amount of supply that must be shifted among allocated inventory proportional to a total amount of allocated inventory that each demand node is able to shift.

21. A computer program product embedded in a non-transitory computer readable medium comprising instructions executable by a computer processor for ad placement planning, the instructions being executable by a computer processor to execute processing comprising:

allocating an inventory of advertising impressions from a plurality of supply nodes to satisfy a demand for advertising impressions received from a plurality of demand nodes, wherein each demand node receives an allocation from at least some of the plurality of supply nodes, and wherein the inventory is allocated in accordance with advertiser targeting criteria associated with each demand node, respectively;
determining a reserve inventory of unallocated advertising impressions for each demand node such that, for each supply node, unallocated advertising impressions are partitioned for allocation among the demand nodes receiving an allocation from a supply node proportionally to the supply node's allocated inventory; and
allocating, for each new demand node, an additional inventory of advertising impressions, wherein the additional inventory satisfies advertiser targeting criteria associated with a new demand node and also satisfies advertiser targeting criteria associated with at least one other demand node, by shifting supply to accommodate the new demand node using a combination of unallocated inventory and shifted inventory, wherein the shifted inventory is shifted out of at least one other demand node's allocation, and wherein the shifted inventory is replaced by the reserve inventory partitioned for allocation to the at least one other demand node.
Patent History
Publication number: 20150348086
Type: Application
Filed: Aug 12, 2015
Publication Date: Dec 3, 2015
Applicant: YAHOO! INC. (Sunnyvale, CA)
Inventors: Erik Vee (San Mateo, CA), Dongni Chen (Sunnyvale, CA), Peiji Chen (San Jose, CA), Satyen Kale (New York, NY), Srinath Mandalapu (Santa Clara, CA), Chandrashekhar Nagarajan (Santa Clara, CA)
Application Number: 14/824,819
Classifications
International Classification: G06Q 30/02 (20060101);