WO2017028693A1

WO2017028693A1 - Bidding control method and apparatus

Info

Publication number: WO2017028693A1
Application number: PCT/CN2016/093890
Authority: WO
Inventors: 魏聪
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2015-08-17
Filing date: 2016-08-08
Publication date: 2017-02-23
Also published as: CN106469384A

Abstract

A bidding control method and apparatus. The method comprises: acquiring a current effect index of target information in bidding words according to bidding information of the target information in the bidding words (S101); if the current effect index is lower than an expected effect index of the target information in the bidding words, acquiring an optimal bidding rank that enables the target information to reach the expected effect index in the bidding words (S102); and bidding for the bidding words according to a bidding price corresponding to the optimal bidding rank (S103). By acquiring a current investment effect of target information and presetting an expected investment effect, an optimal bidding rank that enables the target information to reach the expected investment effect is automatically calculated only when the current investment effect does not reach the expected investment effect, and bidding for bidding words is then performed according to a bidding price corresponding to the optimal bidding rank, such that minimal price adjustment only needs to be carried out while ensuring an advertising effect, and thus the lowest investment cost is ensured.

Description

Auction control method and device

The present application claims priority to Chinese Patent Application Serial No. No. No. No. No. No. No. No. No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No

Technical field

The present application relates to the field of computers, and in particular, to a bidding control method and device.

Background technique

With the widespread use of computer and Internet technologies, more and more users are choosing e-commerce as part of their lifestyle shopping. At the same time, many new technology business areas different from traditional industries have gradually developed along with the rapid development of the Internet. Internet advertising is one of them.

At present, bidding advertising has become one of the most important forms of Internet advertising today. The traditional bidding method requires manual adjustment based on experience, or the advertiser bids based on the budget of the advertiser. Manual price adjustment is acceptable when bidding words are small, but at the same time, when bidding for a large number of bidding words, a lot of labor costs are required. Because of the price adjustment based on experience, it is difficult to reduce the delivery while maximizing the effectiveness of advertising. cost. In another way, since the advertiser bids according to the budget of the advertiser, the user's consumption and click rate are more concerned, and the price cannot be adjusted in real time, which is not conducive to reducing the cost of placing the advertiser.

Summary of the invention

The purpose of the present application is to provide a bidding control method and device for solving the problem of high advertising cost of advertising in the prior art.

To achieve the above objective, the present application provides a bid control method, which includes:

Obtaining a current performance indicator of the target information in the bidding word according to the bidding information in the bidding word according to the target information;

If the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, obtaining an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word;

The bidding word is bidding according to the bid price corresponding to the optimal bidding ranking.

Further, the current performance indicators of the target information in the bidding words are obtained according to the bidding information in the bidding words according to the target information, including:

Obtaining an actual ranking of the target information in the search result of the bidding word according to the bidding information;

Obtaining the current effect index according to the correspondence between the preset ranking and the performance indicator and the actual ranking Standard.

Further, the actual ranking of the target information in the search result of the bidding word is obtained according to the bidding information, including:

a sum of the natural ranking of the target information in the search result of the bidding word and the bidding information of the target information participating in the bidding as a natural ranking of the target information in the search result of the bidding word, wherein The target information has not participated in the bidding of the bidding word, and the bidding information includes a natural ranking of the target information in the search result of the bidding word and a bidding information after the target information participates in the bidding;

Obtain current performance indicators according to the corresponding relationship between the preset ranking and the performance indicator and the actual ranking, including:

Obtaining, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the natural actual ranking, and using the actual performance index corresponding to the natural actual ranking as the current performance indicator.

And a sum of the natural ranking and the bidding information of the target information in the search result of the bidding word as a natural actual ranking of the target information in the search result of the bidding word, wherein the target information has participated in the The bidding of the bidding word, the bidding information includes a natural ranking of the target information in the search result of the bidding word, a current bidding ranking of the target information in the bidding word, and a bidding information number;

The current bid position of the target information in the bid word is used as the actual ranking of the bid of the target information in the search result of the bid word;

Obtaining, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the actual ranking of the natural and the actual ranking of the bid, and the effect corresponding to the actual ranking of the actual ranking of the natural ranking The sum of the indicators serves as the current performance indicator.

Further, obtaining an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word, including:

Determining the difference between the desired effect indicator and the actual performance index corresponding to the natural actual ranking as an optimal effect indicator;

Obtaining an optimal competition corresponding to the optimal performance indicator according to the correspondence between the ranking and the performance indicator Price ranking.

When the natural ranking of the target information in the search result of the bidding word is less than or equal to the preset ranking, the actual ranking of the target information in the search result of the bidding word is obtained according to the bidding information.

When the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking, the current bidding ranking of the target information in the bidding word is used as the bidding of the target information in the search result of the bidding word. The actual ranking, wherein the bidding information includes a natural ranking of the target information in the search result of the bidding word, a current bidding ranking of the target information in the bidding word, and a bidding information number;

Obtaining an actual ranking of the target information in the search result of the bidding word according to the bidding information, including:

And obtaining, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the actual ranking of the bid, and using an effect indicator corresponding to the actual ranking of the bid as the current performance indicator.

Using the desired effect indicator as an optimal effect indicator;

Obtaining an optimal bidding ranking corresponding to the optimal performance indicator according to a correspondence between the ranking and the performance indicator.

Further, bidding the bidding words according to the bid price corresponding to the optimal bidding ranking, including:

If the bid price corresponding to the optimal bid position is lower than the budget threshold, bidding the bid price on the bid price corresponding to the optimal bid position;

If the bid price corresponding to the optimal bid position is equal to or higher than the budget threshold, the bid word is bid by the budget threshold.

Further, the performance indicator includes a click pass rate.

According to another aspect of the present application, there is also provided an auction control device, the device comprising:

An effect obtaining device, configured to acquire, according to the bidding information in the bidding word, the current performance indicator of the target information in the bidding word;

And an optimization device, configured to obtain an optimal result index that achieves the desired target information in the bidding word if the current performance indicator is lower than an expected performance indicator of the target information in the bidding word ranking of bidding;

a bidding device, configured to bid on the bidding word according to an auction price corresponding to the optimal bidding ranking.

Further, the effect obtaining device includes:

a ranking obtaining module, configured to obtain, according to the bidding information, an actual ranking of the target information in a search result of the bidding word;

The effect calculation module is configured to obtain a current performance indicator according to the correspondence between the preset ranking and the performance indicator and the actual ranking.

Further, the ranking obtaining module is configured to use, as the target information, the bidding term, the sum of the natural ranking of the target information in the search result of the bidding word and the bidding information of the target information participating in the bidding The actual ranking of the natural search results, wherein the target information has not participated in the bidding of the bidding words, the bidding information including the natural ranking of the target information in the search results of the bidding words and the assumption of the target information The number of bidding information after participating in the auction;

The effect calculation module is configured to obtain an effect indicator corresponding to the natural actual ranking according to the correspondence between the ranking and the effect indicator, and use the effect indicator corresponding to the natural actual ranking as the current effect indicator.

Further, the ranking obtaining module is configured to use a sum of a natural ranking and a bidding information of the target information in the search result of the bidding word as a natural reality of the target information in the search result of the bidding word. a ranking, wherein the target information has participated in the bidding of the bidding word, the bidding information includes a natural ranking of the target information in the search result of the bidding word, a current bidding ranking of the target information in the bidding word, and an auction a number of information; and a current bid position of the target information in the bid word as an actual ranking of the bid of the target information in the search result of the bid word;

The effect calculation module is configured to respectively obtain an effect indicator corresponding to the actual position of the natural and the actual position of the bid according to the correspondence between the ranking and the performance indicator, and compare the effect indicator corresponding to the actual actual ranking The sum of the performance indicators corresponding to the actual ranking of the bid is used as the current performance indicator.

Further, the optimizing apparatus is configured to: if the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, corresponding to the expected actual performance indicator and the natural actual ranking The difference between the performance indicators is the optimal performance indicator; and the corresponding relationship between the ranking and the performance indicator is obtained. Taking the optimal bidding ranking corresponding to the optimal performance indicator.

Further, the ranking obtaining module is configured to: when the natural ranking of the target information in the search result of the bidding word is less than or equal to the preset ranking, obtain the search for the target information in the bidding word according to the bidding information. The actual ranking in the results.

Further, the ranking obtaining module is configured to use the current bidding ranking of the target information in the bidding word as the target information when the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking. An actual ranking of bids in the search results of the bid words, wherein the bid information includes a natural ranking of the target information in the search results of the bid words, a current bid ranking of the target information in the bid words, and Number of bidding information;

The effect calculation module is configured to obtain an effect indicator corresponding to the actual ranking of the bid according to the correspondence between the ranking and the performance indicator, and use an effect indicator corresponding to the actual ranking of the bid as the current effect indicator.

Further, the optimizing apparatus is configured to: if the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, use the expected effect indicator as an optimal effect indicator; Obtaining an optimal bidding ranking corresponding to the optimal performance indicator according to a correspondence between the ranking and the performance indicator.

Further, the bidding device is configured to: if the bid price corresponding to the optimal bid position is lower than a budget threshold, bid on the bid word corresponding to the bid price corresponding to the optimal bid position; If the bid price corresponding to the optimal bid position is equal to or higher than the budget threshold, the bid word is bidding with the budget threshold.

Further, the performance indicator includes a click pass rate.

Compared with the prior art, the current delivery effect of the target information is calculated by the bid information of the target information in the current bid word, and the expected effect of the advertisement target on the target information is preset by the expected performance indicator, only When the current delivery effect does not reach the expected delivery effect, the optimal bidding ranking that makes the target information reach the expected delivery effect is automatically calculated, and then the bidding price is bidding according to the bidding price corresponding to the optimal bidding ranking, so While ensuring the effectiveness of the ad delivery, only a minimum price adjustment is required to ensure the lowest delivery cost.

In addition, when calculating the performance of the advertisement, the combination of the natural ranking and the auction ranking is adopted, and the effect of the advertisement that can be achieved by the natural ranking is also considered, which can reduce the cost input of the advertiser in improving the bidding ranking, and further reduce The cost of delivery.

DRAWINGS

Other features, objects, and advantages of the present application will become more apparent from the detailed description of the accompanying drawings.

FIG. 1 is a flowchart of a bid control method according to an embodiment of the present application;

2 is a flowchart of a preferred bid control method according to an embodiment of the present application;

Figure 3 (a) and (b) are schematic diagrams of search results before and after bidding for a certain target information using the bid control method provided by the present application;

4(a) and (b) are schematic diagrams showing search results of a target information before and after adjusting the bid price by using the bid control method provided by the present application;

FIG. 5 is a flowchart of another preferred bid control method according to an embodiment of the present application;

FIG. 6 is a schematic structural diagram of a bidding control apparatus according to an embodiment of the present application;

FIG. 7 is a schematic structural diagram of an effect obtaining apparatus in an auction control device according to an embodiment of the present application;

FIG. 8 is a schematic structural diagram of another auction control device according to an embodiment of the present disclosure;

The same or similar reference numerals in the drawings denote the same or similar components.

detailed description

The present application is further described in detail below with reference to the accompanying drawings.

In a typical configuration of the present application, the terminal, the device of the service network, and the trusted party each include one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory. Memory is an example of a computer readable medium.

Computer readable media includes both permanent and non-persistent, removable and non-removable media. Information storage can be implemented by any method or technology. The information can be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, A magnetic tape cartridge, magnetic tape storage or other magnetic storage device or any other non-transportable medium that can be used to store information that can be accessed by a computing device. Computer readable media is not included as defined herein Non-transitory computer readable media (transitory media), such as modulated data signals and carrier waves.

FIG. 1 shows a bid control method provided by an embodiment of the present application, which method includes the following steps:

Step S101: Obtain, according to the bid information of the target information in the bidding word, the current performance indicator of the target information in the bidding word;

Step S102: If the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, obtain an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word. ;

Step S103, bidding the bidding words according to the bid price corresponding to the optimal bidding ranking.

Here, the target information refers to information that the advertisement stalker needs to advertise, and may be a specific web page, a product, an information item, or the like. The bidding word refers to the keyword of the bidding advertisement. When a user searches for a keyword using the search engine, the search result will display all the target information that has already placed the bidding advertisement in the keyword. The current performance indicator and the desired performance indicator are respectively used to indicate the current delivery effect of the target information and the expected delivery effect that the advertisement producer hopes to achieve the target information, and the CTR (Click Through Rate) can be adopted in practical applications. )Wait. In this embodiment, the current delivery effect of the target information is calculated by using the bid information of the target information in the current bid word, and the expected effect of the advertisement target on the target information is preset by the expected performance indicator, only when the current delivery is performed. When the effect does not reach the expected delivery effect, the optimal bidding ranking that makes the target information reach the expected delivery effect is automatically calculated, and then the bidding word is bidding according to the bid price corresponding to the optimal bidding ranking, so the advertising is guaranteed. At the same time, only a minimum price adjustment is required to ensure the lowest cost of delivery.

Further, the embodiment of the present application further provides a preferred bid control method. Referring to FIG. 1, the method specifically includes the steps shown in FIG. 2:

Step S1011: Obtain an actual ranking of the target information in the search result of the bidding word according to the bidding information.

In step S1012, the current effect indicator is obtained according to the correspondence between the preset ranking and the effect indicator and the actual ranking.

Here, the actual ranking of the target information in the search result of the bidding word refers to the position actually displayed by the target information in the search result. The correspondence between the ranking and the performance indicator indicates a correspondence between the ranking of the target information in the search result and the advertising effect of the target information, and is generally set in a preset manner. Taking the click rate as an example, the corresponding relationship between ranking and performance indicators is shown in Table 1:

Ranking

Click rate

11	35％35%
22	29％29%
33	16％16%
44	10％10%
55	7％7%
66	5％5%
77	4％4%
88	3.5％3.5%
99	3％3%
1010	2.5％2.5%
……......	……......

Table 1

In practical applications, the correspondence between ranking and click-through rate can be obtained through long-term data monitoring and summary, and can be adjusted according to subsequent data. In addition, the correspondences are different for different bidding words or different search engines.

In order to advertise the target information A webpage, assuming that the A webpage is related to the food, the advertiser needs to place the bidding advertisement on the A webpage in the bidding words such as "Gourmet" and "Szechuan". When a user searches using the above bidding words, the search results generally have two parts, including a promotion area displaying the bidding advertisement and a natural result area displaying the natural search result, and the ranking of the bidding advertisement in the search result is generally displayed before the natural search result. To achieve the purpose of advertising. In the bidding, there may be two situations: First, the target information has not yet participated in the bidding of a certain bidding word. At this time, there is no such target information in the promotion area of the search result; Second, the target information has already participated in a certain bidding word. Bidding, at this time, the target information is already available in the promotion area of the search results.

For the first case, since the target information has not participated in the bidding of the bidding words, the target information is not present in the promotion area in the search result, and the target information is only displayed in the natural result area. At this time, the bidding information includes a natural ranking of the target information in the search result of the bidding word and a bidding information after the target information participates in the bidding, wherein the bidding ranking of the target information is the target information in the promotion area. Ranking, the natural ranking of the target information is the ranking of the target information in the natural result area. Taking the aforementioned A website as an example, FIG. 3(a) shows the search result after the user searches the bidding word “Szechuan” using the search engine, wherein the promotion area is the area 3A and the natural result area is the area 3B. The number of information displayed in the promotion area 3A is the current number of bidding information, indicating the number of target information currently participating in the bidding bid.

In this case, the step S1011 specifically includes: searching for the target information in the bidding word. The sum of the natural ranking in the result and the number of bidding information of the target information participating in the bidding is used as the natural actual ranking of the target information in the search result of the bidding word. Correspondingly, the step S1012 specifically includes: obtaining an effect indicator corresponding to the natural actual ranking according to the correspondence between the ranking and the effect indicator, and using the actual performance index corresponding to the natural actual ranking as the current Performance indicator.

For example, the auction information in the auction title “Szechuan” is still taken as an example. The natural ranking n in the search results is 3, and the number of bidding information m currently participating in the auction is 3, because when the A website participates in the auction, the bid information The number will be automatically increased by 1, so the number of bidding information m' after the target information participates in the bidding is m+1=4, and FIG. 3(b) shows a schematic diagram of the search results that may occur after the A website participates in the bidding. Since the bidding information is not currently available, the bidding information may not be obtained from the current bidding ranking, or an invalid current bidding rank c may be obtained (for example, the value of c is null or 0, so that the current bidding ranking has no corresponding performance indicator). The ranking of the promotion area information in the search results is displayed in front of the information in the natural result area, so the natural actual ranking of the A website is n+m'=7. Based on the correspondence between the ranking and the performance indicator shown in Table 1, the click rate corresponding to the actual actual ranking is obtained = f(n+m')=f(7)=4%, where f() represents The ranking obtains the processing function of the performance indicator, and thus the target information "A website" has a current hit rate of 4% in the bidding word "Szechuan".

For the second case, since the target information has already participated in the bidding of the bidding word, the target information is displayed in the promotion area and the natural result area in the search result. At this time, the bidding information includes a natural ranking of the target information in the search result of the bidding word, a current bidding ranking of the target information in the bidding word, and a bidding information number, wherein the bidding ranking of the target information is the target information. In the ranking of the promotion area, the natural ranking of the target information is the ranking of the target information in the natural result area. Taking the aforementioned A website as an example, FIG. 4(a) shows the search result after the user searches the bidding word “gourmet” using the search engine, wherein the promotion area is the area 4A and the natural result area is the area 4B. The number of information displayed in the promotion area 4A is the current number of bidding information, indicating the number of target information currently participating in the bidding bid. In practical applications, the natural ranking of the target information in the search results of the bidding words can be automatically obtained through a background program (such as a web crawler), and the number of bidding information and the current bidding ranking can be provided through an advertiser's bidding platform. (Application Programming Interface, API).

In this case, the step S1011 specifically includes: using the sum of the natural ranking and the bidding information of the target information in the search result of the bidding word as the target information in the search result of the bidding word. The actual ranking of the natural; and the current bidding ranking of the target information in the bidding word as the actual ranking of the bidding of the target information in the search results of the bidding word. Correspondingly, the step S1012 includes: separately acquiring the natural actual ranking and competition according to the correspondence between the ranking and the performance indicator. The actual index corresponding to the actual ranking of the price, and the sum of the performance index corresponding to the actual ranking of the natural ranking and the actual ranking of the bid is used as the current performance indicator.

For example, the auction information in the auction word "Gourmet" of the A website is taken as an example. The natural ranking n in the search results is 2, and the number of bidding information m including the A website is 4, in all the information participating in the bidding. A website current bid position c is 4. Since the A website currently participates in the auction of the bidding word “Gourmet”, the number of bidding information will not be changed after the re-bidding, so the actual actual ranking of the website A in the search results of the bidding words is n+m. =6, and the actual ranking of the bids of the A website in the search results of the bid words is the current bid position c, and FIG. 4(b) shows a schematic diagram of the search results that may occur after the website A adjusts the bid price. Based on the correspondence between the ranking and the performance indicator shown in Table 1, the click rate corresponding to the actual actual ranking is obtained = f(n+m)=f(6)=5%, and the actual ranking of the bid is obtained. The corresponding click rate=f(c)=f(4)=10%, so that the current click rate of the target information “A website” in the bidding word “Gourmet” is f(n+m)+f(c). =15%.

Thus, the current CTR algorithm is as follows:

Further, in step S102, obtaining an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word, specifically including:

Determining the difference between the expected performance indicator and the actual performance index corresponding to the natural actual ranking as an optimal performance indicator, and obtaining an optimal corresponding to the optimal performance index according to the correspondence between the ranking and the performance index ranking of bidding.

For example, in the first case described above, the target information "A website" has a desired click rate e of 30% in the bidding word "Szechuan", and its corresponding optimal click rate R is ef (n+). m')=26%, according to the correspondence between ranking and click-through rate, when the ranking is 2, the click-through rate is 29%, and when the ranking is 3, the click-through rate is 16%, so in order to achieve the optimal performance index of 26%, The optimal bid price is 2. Then, the bidding word can be bidding according to the bid price corresponding to the optimal bidding ranking.

For another example, in the foregoing second case, since the actual ranking of the bidding of the target information in the promotion area in the search result changes after the bid adjustment, the performance index corresponding to the actual ranking of the bid will also change, so When obtaining an optimal performance indicator, only the difference between the expected performance indicator and the actual performance index corresponding to the natural actual ranking may be considered. Assuming that the desired click rate e of the target information "A website" in the bidding word "Gourmet" is set to 20%, its corresponding optimal click rate R is e-f(n+m)=15%. According to the row The corresponding relationship between the name and the click rate, when the ranking is 3, the click rate is 16%, and when the ranking is 4, the click rate is 10%, so in order to achieve the optimal performance index of 15%, the optimal bid position is 3. Then, the bidding word can be bidding according to the bid price corresponding to the optimal bidding ranking.

Since the combination of natural ranking and bidding ranking is used in calculating the performance of the advertisement, the effect of the advertisement that can be achieved by the natural ranking is also considered, which can reduce the cost input of the advertiser in improving the auction ranking, and further reduce the delivery. cost.

In addition, for the advertisers, it is also necessary to consider the cost of advertising when bidding, and some popular bidding words, the top ranking of the auction ranking may be very high, and the delivery cost is limited. Under the premise, if a higher cost is invested in a few keywords, the overall delivery effect may be reduced. To this end, in step S103, bidding the bidding word according to the bid price corresponding to the optimal bidding rank may specifically include: if the bid price corresponding to the optimal bidding rank is lower than a budget threshold, The bid price corresponding to the preferred bidding price bids on the bidding word; if the bid price corresponding to the optimal bidding rank is equal to or higher than the budget threshold, the bidding word is bidding with the budget threshold. By setting a budget threshold for the bid price of the bidding word, the bid price of each bid word is limited, the investment in each bid word is reasonably allocated, and the overall effect of the advertisement is improved.

Further, the embodiment of the present application provides another preferred bid control method, where the step S1011 includes: when the natural ranking of the target information in the search result of the bid word is less than or equal to the preset rank. Obtaining, according to the bidding information, an actual ranking of the target information in the search result of the bidding word. Since in actual applications, the search results often contain a lot of information, only a certain amount of information will be displayed on the first page of the search results, and the information of the promotion area will be displayed on the first page, and the information of the natural result area. Except for the first page that can display a portion, the rest will be displayed in subsequent pages by ranking. For the information displayed on subsequent pages, the click rate is often smaller than the information displayed on the first page, and the lower the number of pages, the lower the click rate. Therefore, in the actual processing, only the information displayed on the first page or the first few pages can be considered, which can be achieved by setting a preset ranking. For example, if only 10 natural search results are displayed in the first page of a search engine, the information with a natural rank greater than 10 will be displayed on the second page, and the preset rank can be set to 10. The processing of the subsequent steps S1012, S102, and S103 is performed only when the natural ranking of the target information in the search result of the bidding word is less than or equal to 10.

In addition, the present application also provides an auction control method suitable for another situation, where the bidding control is performed when the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking. The processing flow of the method is as shown in FIG. 5, including:

Step S1011', when the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking, the current bidding ranking of the target information in the bidding word is used as the searching of the target information in the bidding word. The actual ranking of the bids in the results, wherein the bidding information includes a natural ranking of the target information in the search results of the bidding words, a current bidding ranking of the target information in the bidding words, and a number of bidding information.

In step S1012, an effect indicator corresponding to the actual ranking of the bid is obtained according to the correspondence between the ranking and the performance indicator, and an effect indicator corresponding to the actual ranking of the bid is used as the current effect indicator.

Step S102 ′, if the current performance indicator is lower than the expected performance index of the target information in the bidding word, acquiring an optimal bid for achieving the desired performance index of the target information in the bidding word. Ranking.

Step S103', bidding the bidding words according to the bid price corresponding to the optimal bidding ranking.

Since the natural ranking is greater than the preset ranking, the click rate that can be generated is small, so it can be directly ignored to improve the processing efficiency. When calculating the current clickthrough rate, you can only consider the clickthrough rate generated by its current auction position. Correspondingly, obtaining an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word in step S102', specifically comprising: using the expected performance indicator as an optimal performance indicator, according to the ranking and Corresponding relationship of the effect indicators obtains an optimal bidding ranking corresponding to the optimal effect indicator.

For example, taking the bidding information of the A website in the bidding word "hot pot" as an example, and using the click rate as the performance index, the natural ranking n of the website A in the search results is 38, and the number of bidding information including the A website is m. 4. In all of the information participating in the auction, the current bid position c of the website A is 4, and the preset position is set to 10. Since the natural ranking of the A website in the search result of the bidding word "hot pot" is larger than the preset ranking, its current click rate CTR is f(c)=10%. If the expected click rate e is 25%, since CTR < e, the bid price needs to be adjusted. Further, the optimal click rate R=e, as shown in Table 1, the optimal bid position is 2. Therefore, the bid price is bidding according to the bid price corresponding to the optimal bidding ranking, and the purchase price is adjusted.

FIG. 6 shows an auction control device provided by an embodiment of the present application, which includes an effect acquisition device, an optimization device, and a bidding device. Specifically, the effect obtaining device 610 is configured to obtain, according to the bid information in the bidding word, the current performance indicator of the target information in the bidding word; the optimizing device 620 is configured to use the current effect. The indicator is lower than the desired effect of the target information in the bidding word And obtaining an optimal bidding ranking for achieving the desired performance indicator in the bidding word; the bidding device 630 is configured to perform the bidding word according to the bid price corresponding to the optimal bidding ranking bid.

Here, those skilled in the art should understand that the device may include, but is not limited to, an implementation such as a network host, a single network server, a plurality of network server sets, or a cloud computing based computer set. Here, the cloud is composed of a large number of host or network servers based on Cloud Computing, which is a kind of distributed computing, a virtual computer composed of a group of loosely coupled computers. Preferably, the device may also be software running on a network host, server or computer.

Further, the embodiment of the present application further provides a preferred bidding control device. Referring to FIG. 6 , the specific structure of the effect obtaining device 610 is as shown in FIG. 7 , and includes a ranking obtaining module 611 and an effect calculating module 612 . Specifically, the ranking obtaining module 611 is configured to obtain an actual ranking of the target information in the search result of the bidding word according to the bidding information; the effect calculating module 612 is configured to use the preset ranking and the performance indicator according to the preset The corresponding relationship and the actual ranking obtain the current performance indicator.

Here, the actual ranking of the target information in the search result of the bidding word refers to the position actually displayed by the target information in the search result. The correspondence between the ranking and the performance indicator indicates a correspondence between the ranking of the target information in the search result and the advertising effect of the target information, and is generally set in a preset manner. Taking the click rate as an example, the correspondence between the ranking and the effect indicator is shown in Table 1.

In practical applications, the correspondence between ranking and click-through rate can be obtained through long-term data monitoring and summary, and can be adjusted according to subsequent data. Also, for different bidding words or different searches The engine, the correspondence is also different.

In this case, the ranking obtaining module 611 is specifically configured to use, as the target information, the sum of the natural ranking of the target information in the search result of the bidding word and the bidding information after the target information participates in the bidding. The actual actual ranking in the search results of the bidding words. Correspondingly, the effect calculation module 612 is specifically configured to acquire an effect indicator corresponding to the natural actual ranking according to the correspondence between the ranking and the effect indicator, and use the effect indicator corresponding to the natural actual ranking as the current Performance indicator.

For example, the auction information in the auction title “Szechuan” is still taken as an example. The natural ranking n in the search results is 3, and the number of bidding information m currently participating in the auction is 3, because when the A website participates in the auction, the bid information The number will be automatically increased by 1, so the number of bidding information m' after the target information participates in the bidding is m+1=4, and FIG. 3(b) shows a schematic diagram of the search results that may occur after the A website participates in the bidding. Since the bidding information is not currently available, the bidding information may not be obtained from the current bidding ranking, or an invalid current bidding rank c may be obtained (for example, the value of c is null or 0, so that the current bidding ranking has no corresponding performance indicator). The ranking of the promotion area information in the search results is displayed in front of the information in the natural result area, so the natural actual ranking of the A website is n+m'=7. Obtaining the natural reality based on the correspondence between the ranking and the performance indicator shown in Table 1 The corresponding click rate = f(n+m')=f(7)=4%, where f() represents the processing function for obtaining the effect indicator by ranking, and thus the target information "A website" is in the bidding word "Szechuan" The current hit rate is 4%.

In this case, the ranking obtaining module 611 is specifically configured to use the sum of the natural ranking and the bidding information of the target information in the search result of the bidding word as the target information in the search result of the bidding word. The actual ranking of the natural; and the current bidding ranking of the target information in the bidding word as the actual ranking of the bidding of the target information in the search results of the bidding word. Correspondingly, the effect calculation module 612 is specifically configured to acquire, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the actual actual ranking and the actual ranking of the bid, and corresponding to the natural actual ranking. The sum of the performance indicators corresponding to the actual rankings of the bids is used as the current performance indicator.

Thus, the current CTR algorithm is as follows:

Further, the optimization device 620 is specifically configured to: if the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, the expected performance indicator and the natural actual ranking The difference between the corresponding performance indicators is used as an optimal performance indicator, and the optimal bidding ranking corresponding to the optimal performance indicator is obtained according to the correspondence between the ranking and the performance indicator.

For example, in the first case described above, the target information "A website" has a desired click rate e of 30% in the bidding word "Szechuan", and its corresponding optimal click rate R is ef (n+). m')=26%, according to the correspondence between ranking and click-through rate, when the ranking is 2, the click-through rate is 29%, and when the ranking is 3, the click-through rate is 16%, so in order to achieve the optimal performance index of 26%, The optimal bid price is 2. Then, the bidding device 630 can bid on the bidding word according to the bid price corresponding to the optimal bidding ranking.

For another example, in the foregoing second case, since the actual ranking of the bidding of the target information in the promotion area in the search result changes after the bid adjustment, the performance index corresponding to the actual ranking of the bid will also change, so When obtaining an optimal performance indicator, only the difference between the expected performance indicator and the actual performance index corresponding to the natural actual ranking may be considered. Assuming that the desired click rate e of the target information "A website" in the bidding word "Gourmet" is set to 20%, its corresponding optimal click rate R is e-f(n+m)=15%. According to the correspondence between ranking and click rate, when the ranking is 3, the click rate is 16%, and when the ranking is 4, the click rate is 10%, so in order to achieve the optimal performance index of 15%, the optimal bid is ranked 3. Then, the bidding device 630 can bid on the bidding word according to the bid price corresponding to the optimal bidding ranking.

In addition, for the advertisers, it is also necessary to consider the cost of advertising when bidding, and some popular bidding words, the top ranking of the auction ranking may be very high, and the delivery cost is limited. Under the premise, if a higher cost is invested in a few keywords, the overall delivery effect may be reduced. To this end, the bidding device 630 bids the bidding words according to the bid price corresponding to the optimal bidding ranking. Specifically, the bidding device 630 is configured to: if the bid price corresponding to the optimal bid position is lower than a budget threshold, bid on the bid price corresponding to the bid price corresponding to the optimal bid position; If the bid price corresponding to the superior bidding rank is equal to or higher than the budget threshold, the bidding term is used for the bidding term Make a bid. By setting a budget threshold for the bid price of the bidding word, the bid price of each bid word is limited, the investment in each bid word is reasonably allocated, and the overall effect of the advertisement is improved.

Further, the embodiment of the present application provides another preferred bidding control device, where the ranking obtaining module 611 is configured to use when the natural ranking of the target information in the search result of the bidding word is less than or equal to the preset ranking. Obtaining, according to the bidding information, an actual ranking of the target information in the search result of the bidding word. Since in actual applications, the search results often contain a lot of information, only a certain amount of information will be displayed on the first page of the search results, and the information of the promotion area will be displayed on the first page, and the information of the natural result area. Except for the first page that can display a portion, the rest will be displayed in subsequent pages by ranking. For the information displayed on subsequent pages, the click rate is often smaller than the information displayed on the first page, and the lower the number of pages, the lower the click rate. Therefore, in the actual processing, only the information displayed on the first page or the first few pages can be considered, which can be achieved by setting a preset ranking. For example, if only 10 natural search results are displayed in the first page of a search engine, the information with a natural rank greater than 10 will be displayed on the second page, and the preset rank can be set to 10. The corresponding processing is performed by the effect calculation module 612, the optimization device 620, and the bidding device 630 only when the natural ranking of the target information in the search results of the bidding words is less than or equal to 10.

In addition, the present application also provides an auction control device suitable for another situation, for performing bid control when the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking. The device structure is as shown in Fig. 8, and includes a ranking obtaining module 611', an effect calculating module 612', an optimizing device 620', and a bidding device 630'. Specifically, the ranking obtaining module 611' is configured to use the current bidding ranking of the target information in the bidding word as the target when the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking. The actual ranking of the bids in the search results of the bidding words, wherein the bidding information includes a natural ranking of the target information in the search results of the bidding words, and a current bidding ranking of the target information in the bidding words. And the number of auction information. The effect calculation module 612 ′ is configured to obtain an effect indicator corresponding to the actual ranking of the bid according to the correspondence between the ranking and the performance indicator, and use an effect indicator corresponding to the actual ranking of the bid as the current performance indicator. . The optimizing device 620' is configured to acquire, if the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, obtain an effect indicator that achieves the desired target information in the bidding word The optimal bidding ranking. The bidding device 630' is configured to bid on the bidding words according to the bid price corresponding to the optimal bidding ranking.

Since the natural ranking is greater than the preset ranking, the click rate that can be generated is small, so it can be directly ignored to improve the processing efficiency. When calculating the current clickthrough rate, you can only consider the clickthrough rate generated by its current auction position. Correspondingly, the optimization device 620' is specifically configured to: if the current performance index is low And the desired effect indicator is used as an optimal effect indicator according to the expected performance index of the target information in the bidding word; and the optimal effect indicator is obtained according to the correspondence between the ranking and the effect indicator. Optimal bidding ranking.

In summary, the technical solution provided by the embodiment of the present application calculates the current delivery effect of the target information by using the bid information of the target information in the current bid word, and presets the target information of the target by the expected effect indicator. Expected delivery effect, only when the current delivery effect does not reach the expected delivery effect, automatically calculate the optimal bidding ranking that makes the target information reach the expected delivery effect, and then according to the bid price corresponding to the optimal bidding ranking The bidding word is used for bidding, so only the minimum price adjustment is required while ensuring the effectiveness of the advertising, thus ensuring the lowest delivery cost. In addition, when calculating the performance of the advertisement, the combination of the natural ranking and the auction ranking is adopted, and the effect of the advertisement that can be achieved by the natural ranking is also considered, which can reduce the cost input of the advertiser in improving the bidding ranking, and further reduce The cost of delivery.

It should be noted that the present application can be implemented in software and/or a combination of software and hardware, for example, using an application specific integrated circuit (ASIC), a general purpose computer, or any other similar hardware device. In one embodiment, the software program of the present application can be executed by a processor to implement the steps or functions described above. Likewise, the software programs (including related data structures) of the present application can be stored in a computer readable recording medium such as a RAM memory, a magnetic or optical drive or a floppy disk and the like. In addition, some of the steps or functions of the present application may be implemented in hardware, for example, as a circuit that cooperates with a processor to perform various steps or functions.

In addition, a portion of the present application can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide a method and/or technical solution in accordance with the present application. The program instructions that invoke the method of the present application may be stored in a fixed or removable record. The recording medium, and/or transmitted by a stream of data in a broadcast or other signal bearing medium, and/or stored in a working memory of a computer device operating in accordance with the program instructions. Herein, an embodiment in accordance with the present application includes a device including a memory for storing computer program instructions and a processor for executing program instructions, wherein when the computer program instructions are executed by the processor, triggering The apparatus operates based on the aforementioned methods and/or technical solutions in accordance with various embodiments of the present application.

It is obvious to those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the invention is defined by the appended claims instead All changes in the meaning and scope of equivalent elements are included in this application. Any reference signs in the claims should not be construed as limiting the claim. In addition, it is to be understood that the word "comprising" does not exclude other elements or steps. A plurality of units or devices recited in the device claims may also be implemented by a unit or device by software or hardware.

Claims

A bid control method, wherein the method comprises:

Obtaining a current performance indicator of the target information in the bidding word according to the bidding information in the bidding word according to the target information;

If the current performance indicator is lower than an expected performance indicator of the target information in the bidding word, obtaining an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word;

The bidding word is bidding according to the bid price corresponding to the optimal bidding ranking.
The method according to claim 1, wherein the current performance indicator of the target information in the bidding word is obtained according to the bidding information in the bidding word according to the target information, including:

Obtaining an actual ranking of the target information in the search result of the bidding word according to the bidding information;

The current performance indicator is obtained according to the correspondence between the preset ranking and the performance indicator and the actual ranking.
The method according to claim 2, wherein the actual ranking of the target information in the search result of the bidding word is obtained according to the bidding information, including:

a sum of the natural ranking of the target information in the search result of the bidding word and the bidding information of the target information participating in the bidding as a natural ranking of the target information in the search result of the bidding word, wherein The target information has not participated in the bidding of the bidding word, and the bidding information includes a natural ranking of the target information in the search result of the bidding word and a bidding information after the target information participates in the bidding;

Obtain current performance indicators according to the corresponding relationship between the preset ranking and the performance indicator and the actual ranking, including:

Obtaining, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the natural actual ranking, and using the actual performance index corresponding to the natural actual ranking as the current performance indicator.
The method according to claim 2, wherein the actual ranking of the target information in the search result of the bidding word is obtained according to the bidding information, including:

And a sum of the natural ranking and the bidding information of the target information in the search result of the bidding word as a natural actual ranking of the target information in the search result of the bidding word, wherein the target information has participated in the The bidding of the bidding word, the bidding information includes a natural ranking of the target information in the search result of the bidding word, a current bidding ranking of the target information in the bidding word, and a bidding information number;

Using the current bidding ranking of the target information in the bidding word as the target information in the bidding word The actual ranking of the auction in the search results;

Obtain current performance indicators according to the corresponding relationship between the preset ranking and the performance indicator and the actual ranking, including:

Obtaining, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the actual ranking of the natural and the actual ranking of the bid, and the effect corresponding to the actual ranking of the actual ranking of the natural ranking The sum of the indicators serves as the current performance indicator.
The method according to claim 3 or 4, wherein obtaining an optimal bidding ranking for causing the target information to achieve a desired performance indicator in the bidding word comprises:

Determining the difference between the desired effect indicator and the actual performance index corresponding to the natural actual ranking as an optimal effect indicator;

Obtaining an optimal bidding ranking corresponding to the optimal performance indicator according to the correspondence between the ranking and the performance indicator.
The method according to any one of claims 3 to 5, wherein the actual ranking of the target information in the search results of the bidding words is obtained according to the bidding information, including:

When the natural ranking of the target information in the search result of the bidding word is less than or equal to the preset ranking, the actual ranking of the target information in the search result of the bidding word is obtained according to the bidding information.
The method according to claim 2, wherein the actual ranking of the target information in the search result of the bidding word is obtained according to the bidding information, including:

When the natural ranking of the target information in the search result of the bidding word is greater than the preset ranking, the current bidding ranking of the target information in the bidding word is used as the bidding of the target information in the search result of the bidding word. The actual ranking, wherein the bidding information includes a natural ranking of the target information in the search result of the bidding word, a current bidding ranking of the target information in the bidding word, and a bidding information number;

Obtaining an actual ranking of the target information in the search result of the bidding word according to the bidding information, including:

And obtaining, according to the correspondence between the ranking and the performance indicator, an effect indicator corresponding to the actual ranking of the bid, and using an effect indicator corresponding to the actual ranking of the bid as the current performance indicator.
The method of claim 7, wherein obtaining an optimal bidding ranking that causes the target information to achieve a desired performance indicator in the bidding word comprises:

Using the desired effect indicator as an optimal effect indicator;

Obtaining an optimal bid row corresponding to the optimal performance indicator according to the correspondence between the ranking and the performance indicator name.
The method according to any one of claims 1 to 8, wherein the bidding words are bidding according to the bid price corresponding to the optimal bidding ranking, including:

If the bid price corresponding to the optimal bid position is lower than the budget threshold, bidding the bid price on the bid price corresponding to the optimal bid position;

If the bid price corresponding to the optimal bid position is equal to or higher than the budget threshold, the bid word is bid by the budget threshold.
The method according to any one of claims 1 to 9, wherein the effect indicator comprises a click through rate.
An auction control device, wherein the device includes:

An effect obtaining device, configured to acquire, according to the bidding information in the bidding word, the current performance indicator of the target information in the bidding word;

And an optimization device, configured to obtain an optimal result index that achieves the desired target information in the bidding word if the current performance indicator is lower than an expected performance indicator of the target information in the bidding word ranking of bidding;

a bidding device, configured to bid on the bidding word according to an auction price corresponding to the optimal bidding ranking.
The device according to claim 11, wherein the effect obtaining means comprises:

a ranking obtaining module, configured to obtain, according to the bidding information, an actual ranking of the target information in a search result of the bidding word;

The effect calculation module is configured to obtain a current performance indicator according to the correspondence between the preset ranking and the performance indicator and the actual ranking.
The device according to claim 12, wherein the ranking obtaining module is configured to use the sum of the natural ranking of the target information in the search result of the bidding word and the bidding information after the target information participates in the bidding Determining a natural actual ranking of the target information in the search result of the bidding word, wherein the target information has not participated in the bidding of the bidding word, and the bidding information includes the target information in the search result of the bidding word. Natural ranking and the number of bidding information after the target information is assumed to participate in the auction;

The effect calculation module is configured to obtain an effect indicator corresponding to the natural actual ranking according to the correspondence between the ranking and the effect indicator, and use the effect indicator corresponding to the natural actual ranking as the current effect indicator.
The device according to claim 12, wherein the ranking obtaining module is configured to use, as the target information, the sum of the natural ranking and the bidding information of the target information in the search result of the bidding word. a natural ranking in the search results, wherein the target information has participated in the bidding of the bidding word, the bidding information includes a natural ranking of the target information in the search results of the bidding words, and the target information is in the bidding a current bidding ranking in the word and a bidding information number; and a current bidding ranking of the target information in the bidding word as an actual ranking of the bidding of the target information in the search result of the bidding word;

The effect calculation module is configured to respectively obtain an effect indicator corresponding to the actual position of the natural and the actual position of the bid according to the correspondence between the ranking and the performance indicator, and compare the effect indicator corresponding to the actual actual ranking The sum of the performance indicators corresponding to the actual ranking of the bid is used as the current performance indicator.
The apparatus according to claim 13 or 14, wherein said optimizing means is configured to: if said current performance indicator is lower than an expected performance indicator of said target information in said bidding word, said desired The difference between the effect indicator and the effect indicator corresponding to the natural actual ranking is used as an optimal effect indicator; and the optimal bid position corresponding to the optimal effect indicator is obtained according to the correspondence between the ranking and the effect indicator.
The device according to any one of claims 13 to 15, wherein the ranking obtaining module is configured to: when the natural ranking of the target information in the search result of the bidding word is less than or equal to a preset ranking, according to the The bid information obtains the actual ranking of the target information in the search results of the bid words.
The device according to claim 12, wherein the ranking obtaining module is configured to: when the natural ranking of the target information in the search result of the bidding word is greater than a preset ranking, the target information is in the bidding word The current bid ranking is the actual ranking of the bids of the target information in the search results of the bidding words, wherein the bidding information includes a natural ranking of the target information in the search results of the bidding words, and the target information is in the The current auction ranking and the number of auction information in the auction term;

The effect calculation module is configured to obtain an effect indicator corresponding to the actual ranking of the bid according to the correspondence between the ranking and the performance indicator, and use an effect indicator corresponding to the actual ranking of the bid as the current effect indicator.
The device according to claim 17, wherein said optimizing means is configured to: if said current performance indicator is lower than an expected performance indicator of said target information in said bidding word, said desired effect indicator As an optimal performance indicator; and obtaining an optimal bidding ranking corresponding to the optimal performance indicator according to the correspondence between the ranking and the performance indicator.
Apparatus according to any one of claims 11 to 18, wherein said bidding means is for If the bid price corresponding to the optimal bid position is lower than the budget threshold, bidding the bid price on the bid price corresponding to the optimal bid position; and if the bid price corresponding to the optimal bid position is equal to or Above the budget threshold, the bid word is bidd with the budget threshold.
The apparatus according to any one of claims 11 to 19, wherein the effect indicator comprises a click through rate.