JP2002133064A - Method for diagnosing advertisement - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an advertisement diagnosing method for diagnosing an advertisement and precisely grasping the improving method of the advertisement. SOLUTION: The constitution factors of main components are set to be the question items of a questionnaire so that the first and second main components become 'the purchase will of a commodity' and 'enjoyment with respect to the advertisement'. The advertisement is caused to view by objects by extracting them from target objects to whom the commodity is to be sold at random, and the answer of the questionnaire is received. The answer is inputted to a computer, the main components are analyzed and 'the purchase will of the commodity' and 'enjoyment with respect to the advertisement', which are the first and second main components, are obtained. An item for improving 'the purchase will of the commodity' and 'enjoyment with respect to the advertisement' can easily be discriminated from the graph of a peculiar vector in the first and second main components.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an advertisement diagnosis method for diagnosing the effect and effectiveness of an advertisement.

[0002]

2. Description of the Related Art Advertising methods include advertisements on broadcast media such as television and radio, advertisements on newspapers and magazines, advertisements on flyers, catalogs, leaflets, pamphlets, etc., advertisements on websites, signboards, and on trains. Advertisements, advertisements by mail, advertisements by structures and moving objects, etc.
Advertising of the organization itself, products and services is made by various methods using various advertising media.

[0003] Diagnosis, evaluation, and further improvement of whether or not these advertisements are effective and effective are conventionally performed only by empirical methods by advertisement experts.

[0004]

In the conventional diagnosis and improvement of advertisement by experts, not only does it take into account the reaction of customers and consumers who receive the advertisement, but it is also necessary to measure the effectiveness of the advertisement itself. Did not. for that reason,
Improvements in advertising by specialists did not necessarily lead to increased ad liking and increased customer and consumer willingness to buy.

Accordingly, an object of the present invention is to diagnose advertisements,
An object of the present invention is to provide an advertisement diagnosis method capable of accurately grasping an improvement method of an advertisement.

[0006]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention creates a questionnaire in which items serving as constituent factors in diagnosing and evaluating advertisements are questionnaire items. Targeting customers and consumers of goods and services to be advertised, and after having the target audience view the advertisement, conduct the questionnaire survey, input the result to a computer and store it in a memory, The main component analysis is performed by the computer based on the data stored in the computer, and either or both of the purchase motivation and the CM liking are obtained as the main components, and the advertisement is diagnosed based on the obtained data of the main components. I made it. In particular, the questionnaire item is desirably set so that either one or both of the purchase intent and the CM liking are the main components, so that either one of the purchase motive or the CM liking is the first principal component and the other is the other. Is designed as the second principal component, and the purchase intent or / and the CM likability are obtained as the main component. Based on this, the CM likability is increased so that the purchase urge is increased. It is possible to easily seek a measure for improving the advertisement.

Also, the principal component score of each sample is plotted and displayed or printed on a coordinate system in which one of the first and second principal components is a horizontal axis and the other is a vertical axis, and the displayed or printed is obtained. By examining the coefficients of each item constituting the principal component while referring to the items, it is easy to determine what items can be improved to increase the willingness to purchase or / and the CM favorability. Diagnose your ad so you can understand it.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Themes, products and services you want to advertise,. The expected purchase target of the product,. The content of the advertisement is affected by the advertising medium. For example, when a product is a food product and when it is a fixture such as a household electric appliance, the content of appealing to a viewer by an advertisement for the product is different. In the case of food, a typical example would be an advertisement that looks good and makes the viewer think they want to eat. In addition, for home electric appliances and the like, it is desirable to provide an advertisement that reminds the viewer that the usability is likely to be good or the quality is likely to be good. Also, even if the product is food, the content to be appealed by the advertisement is different depending on whether the prospective purchase target is a child or an elderly person. Furthermore, depending on whether the advertising medium is performed on a television, a newspaper, a magazine, or the like, the content of the advertisement is regulated, and the power of appealing (appealing power) of the advertisement also changes.

[0009] As described above, whether or not an advertisement is effective is determined by the above three elements of the advertisement. Therefore, the present invention does not diagnose advertisements based on the experience of experts as in the past, but directly incorporates the opinions and opinions of customers and consumers, such as those who purchase products of the advertisements, and performs principal component analysis to analyze the advertisements. Diagnose, consider whether the advertisement is effective, and guide the improvement of the advertisement. By this diagnosis, it is possible to objectively and quantitatively understand which element of the advertisement has “motivation to purchase the product / service” and “favorableness to the advertisement”, that is, “the product / service”
The constituent factors of the main components are designed as questionnaire items (question items) so that “willingness to purchase the service” and “favorableness for the advertisement” can clearly appear.

[0010] Therefore, the advertisement itself, the production of the advertisement,
The main components such as impressions, images, influences, etc. regarding the product in the advertisement are set as questionnaire survey items.
Further, when the principal component analysis is performed based on the data obtained by this questionnaire, “the willingness to purchase the product / service” and “the favorable impression for the advertisement” appear as the first principal component and the second principal component. The questionnaire items are set taking into account the attributes, behavioral patterns, and intentions of the respondents.

At this time, the answer options for the survey items are:
In order to obtain a quantitative answer, the options are created in a normal order, and the number of the options is increased from a small number to a large number. or,
If the answer is qualitative, “1” is assigned to each option, and “0” is assigned to each alternative. When a plurality of options are ranked and selected, the score is increased from the higher rank. For example, Q of the following questionnaire example
In 14, “2” is selected for the option selected as the first place A, and “2” for the option selected as the second place B.
“1” is assigned, and “0” is assigned to an unselected option. Furthermore, by eliminating neutral options such as "I can't say anything" or "I can't say anything", the variability of answers is increased, and the main component is easily precipitated. In addition to putting heading labels on the survey items, the interpretation of the principal component vector and improvement measures for the commercial were found, such as normalization of qualitative data by [0] and [1] and quantification when there were multiple answers. Input design so that it is easy to do.

[0012] For example, the following is an example of a questionnaire for diagnosing a TV advertisement for this product when the product is food (potato chips). This survey item is used by selecting from a large number of survey items prepared in advance in accordance with the targeted advertisement. Therefore, the survey item numbers are not always consecutive. 1st Stage attribute Q1. Please answer your gender. 1. Male 2. Female <Gender> Q2. Please enter your full age. Age <age> Action pattern Q11. I will ask you how to spend your holiday. <Going out on holidays> 1. Almost never go out 2. Occasionally go out 3. Go out a lot 4. Always go out Q13. How many hours a day do you watch TV on weekdays? <Number of hours watched> 1. Rarely watched 2. About 1 hour 3. About 2 to 3 hours 4. About 4 to 5 hours 5. More than 6 hours Q14. Ask about the types of TV programs you often watch. Please fill in the first place to A and the second place to B in the questionnaire survey. <Program> 1. Animation 2. Variety (comedy program) A = 3. Music 4. Sports 5. Drama 6. Historical drama B = 7.Movies 8.Wide Shows 9.Documents 10.Cooking/Gourmet 11.Travel/Journey/History 12.Quiz 13.Information / News 14.Education/Culture Q15. Ask about when you watch TV often. <Time zone> 1. Almost never 2. 5: 00-8: 00 3. 〜12: 00 A ＝ 4. ~ 16: 00 5. ~ 19: 00 6. ~ 22: 00 7. ~ 24: 00 8. After 24:00 B = Background Q21. What do you think about garbage separation? <Garbage separation> 1. Do not separate. 2. I don't think it makes sense. 3. I don't think there is no way. 4. I think it is natural. 5. I should actively cooperate. Do you often watch variety shows? <Harraity> 1. I do not see it at all 2. I do not see it much 3. I see it sometimes 4. I look closely Q24. Are you a good look at travel, travel and history? <Travel / Journey / History> 1. I do not see it at all. What do you think about fashion items? <Following fashion items> 1. I don't like it 2. I don't care 3. I care to a certain degree 4. I'm a chaseer Q28. What do you think about buzzwords and talent gags?

<Buzzword Reaction> 1. I don't like it 2. I don't like it 3. I care a little 4. I'm a chase 2nd Stage CM Q31. Ask about the characters in this commercial. <CM characters> 1. I don't like it 2. I don't like it so much 3. I like it a little 4. I love Q32. Does the character of this commercial match the product image of this product? <Person / product matching> 1. Not at all 2. Not very well 3. Somewhat agree 4. Quite agree Q33. I'm asking about the commercial slogan, "You can't st op when you open it." <Coffe favorability> 1. I do not like it 2. I do not leave an impression 3. I can have a good feeling 4. I liked it Q38. I'm asking about the number of times this commercial is running. <CM frequency> 1. Rarely seen 2. Not often seen 3. Frequently seen 4. Too persistent Q39. Is this commercial boring? <CM boredom> 1. I'm not interested 2. It's pretty boring 3. It's a bit boring 4. I don't get bored I'm asking about this commercial BGM (music playing in the background). <BGM> 1. I didn't like it 2. I didn't notice 3. I liked it 4. I liked it. I'm asking about the sound effects (how to use sounds) of this commercial. <Sound effects> 1. I didn't like it 2. I didn't notice 3. I liked it 4. I liked Q44. How fast was the screen switching for this commercial? <Screen switching> 1. Slow 2. Just good 3. Fast 4. Intense Q45. How was the color of the screen of this commercial? <Screen color> 1. Displeased 2. I don't like it 3. I like it 4. I like it Q46. How was the scene setting for this commercial? <Scene setting favorability> 1. I do not like it 2. I do not leave an impression 3. I can have a good feeling 4. Favorite product Q51. I'd like to ask about the impression of the product from this commercial.

<Product Impression> 1. I do not like it 2. I do not leave an impression 3. I can have a good feeling 4. Good Q52. Was the number of appearances of the product in this commercial appropriate? <Number of product appearances> 1. Too few 2. Not enough 3. Appropriate 4. Too many Q53. Are you familiar with the products featured in this commercial? 1. I almost don't know 2. I don't know much <Product details> 3. I understand to some extent 4. I understand well General Q61. I'm asking about this whole commercial. <Commercial liking> 1. I don't like it 2. I don't like it 3. I can get some liking 4. I liked Q62. Ask about P _& G's corporate image after seeing this commercial. 1.I got worse 2.I hardly changed 3.I got better <Change in corporate image> 4.I didn't know where the company was. 5.I knew for the first time with this CM <Recognition of company name> Q63. Did anything that touches your heart look at this commercial?

<CM co-sensitivity> 1. I felt repulsion. 2. There was nothing. 3. It resonated a little. 4. Something resonated with my heart. Q65-A. Which of these commercials need improvement? (Multiple answers possible) <Improvement> 1. BGM (background music) 2. Characters 3. Scene settings 4. Narration (narration) 5. Dialogue 6. Screen switching and other video methods 7. None Q65-B. Which of these commercials need improvement? (Multiple answers allowed) 7. None 8. Catch copy 9. Description of product content 10. Explanation of quality such as safety 11. Corporate attitude 12. Environmental consideration of product Q66. Have you seen this commercial and wanted to buy this product? <Motivation to purchase> 1. I do not want to buy 2. I do not feel the need 3. I want to buy if I can afford 4. I want to buy by all means Q67. I'm asking about the quality of this commercial. <CM grade> 1.Significantly lacking grade 2.Slightly lacking grade 3.Dependingly graded 4.Elegant In the example of the questionnaire described above, the components of the main component were designed as question items for each group. I have. Gender (Q1), age (Q2) as “attributes”, viewing hours (Q13), favorite TV programs (Q14),
For "background", garbage separation (Q21), preference for variety programs (Q23), preference for travel, travel, history programs (Q24), response to fashion (Q27, Q28), "commercial" itself, copy Positive (Q3
3) For the boring of commercials (Q39) and “staging”, BGM (Q4
1), the speed of screen switching (Q44), the color of the screen (Q45), the liking of the scene setting (Q46), and the number of product appearances in commercials (Q52), ”As C
M is favorable (Q61), CM's sympathy (Q63), CM's motivation to purchase (Q66), and CM's quality (Q67). Also, <sex>, <age>, <going out of holiday>, <viewing time>,... Described on the right end of the questionnaire are heading labels for identifying question items.

A person who randomly selected the above-mentioned questionnaire from a group expected to be a purchase target (customers and consumers), that is, a group expected to be a purchase target by generation, age, occupation, gender, and the like. Survey and get answers. In this case, a questionnaire is given to the subjects who have more than the number of question items, and samples without missing values (responses to all items) are collected more than the number of question items.

Hereinafter, a description will be given with reference to an example in which the above-described questionnaire is conducted and a TV advertisement is diagnosed. In this example, since the product is potato chips, a student is assumed as a prospective purchaser, and a questionnaire survey is conducted on randomly selected students. Also, in this example, since the advertisement is a television advertisement, the advertisement was viewed several times, and attention was paid so that there was no missing value.

A principal component analysis is performed on the collected sample by a computer. FIG. 1 is an example of a configuration of a computer 10 that performs principal component analysis according to an embodiment of the present invention. In FIG. 1, reference numeral 11 denotes a processor, 12 denotes a ROM in which a system program is stored, 13 denotes a RAM, which is partially constituted by a non-volatile RAM.
A program or the like for performing principal component analysis is stored in advance in the portion of M. Reference numeral 14 denotes a hard disk, and 15 denotes an interface for inputting data and programs from a floppy (registered trademark) disk, a CD, or the like.
Reference numeral 16 denotes a display device such as a CRT or a liquid crystal, and reference numeral 17 denotes data input means constituted by a keyboard, a mouse, and the like. 1
Reference numeral 8 denotes a printer. These elements are connected by a bus 19.

FIG. 2 is a flowchart of a computer process for diagnosing an advertisement based on a questionnaire result.
The questionnaire result data is input by reading the answer from the keyboard or automatically (step S1). At this time, the gender of the question item Q1 is "1" for men and "2" for women.
As the age of the question item Q2, the entered age is input as it is. For the answer to the question item for selecting a quantitative option, the number of the option is input. In the question item for selecting a more qualitative option, "1" is selected for each option, and "0" is not selected for each option.
Enter In addition, the qualitative options that are ranked and selected are “2” for the first option, “1” for the second option, and “0” for the option not selected. ", And input the scores (weights) in the normal order.

As shown in FIG. 3, the data input in this manner is, for each sample, a survey item (a question item for selecting a quantitative option is referred to as a survey item).
In the case of a question item for selecting a qualitative option, the number of scores (weights) of the option itself is referred to as a survey item is stored.

Based on the input data, standardization is performed for each survey item (step S2). First, before standardization, an average value xa and a standard deviation σ for each score xi of each survey item are obtained, and standardized data Z for each survey item is obtained by the following first formula.

Zi = (xi−xa) / σ (1) On the basis of the standardized data Z obtained in this way, a calculation of the following second equation is performed to obtain a correlation coefficient r between the respective survey items. (Step S3).

Assuming that the survey items a and b are survey items, and that the values standardized by the first formula for the survey items a and b of each sample i are Zai and Zbi, respectively, the correlation coefficient rab between the survey items a and b is as follows. Is expressed by the second equation.

[0024]

(Equation 1)

Based on the correlation matrix, principal component analysis is performed to obtain a first principal component A and a second principal component B represented by the following third and fourth equations (step S4).

A = k ₁₁ · (1) + k ₁₂ · (2) + k ₁₃ · (3) + ··· + k _1n · (n) (3) B = k ₂₁ · (1) + k ₂₂ · ( 2) + k _23. (3) +... + K _2n. (N) (4) However, principal components A and B are independent of each other, and k ₁₁ ² + k ₁₂ ² + k ₁₃ ² +. .. + K _1n ² = 1 (5) k ₂₁ ² + k ₂₂ ² + k ₂₃ ² +... + K _2n ² = 1 (6) k ₁₁ k ₂₁ + k ₁₂ k ₂₂ + k ₁₃ k _23. + K _1n k _2n = 0 (7) In the expressions of the first and second principal components, (1), (2),
(3),... (N) represent the standardized data of the survey items, and the coefficients k ₁₁ , k ₁₂ , k _13.
k _1n , k ₂₁ , k ₂₂ , k ₂₃ ... k _2n are the components of the eigenvectors of the first and second principal components and are derived based on the correlation matrix.

Next, contribution rates α and β are obtained from the eigenvalues λ _a and λ _{b of} the first principal component A and the second principal component B (step S).
5). In the principal component analysis, the number n of the eigenvectors of the correlation matrix matches the smaller number of the survey item and the number of samples. Λ ₁ , λ ₂ , λ ₃ ,... Λ _n
Then, theoretically, λ ₁ > λ ₂ > λ ₃ >...> Λ _n > 0
Λ ₁ + λ ₂ + λ ₃ ... + Λ _n = n, where (λ _i / n) is called the contribution ratio of the _i- th principal component,
Usually expressed as a percentage (%). And the contribution rate (λ _i / n)
Are the first principal component, the second principal component,...
The greater the contribution (λ _i / n), the greater the influence,
In the present invention, it means that the persuasiveness of the advertisement is high.

Then, it is determined whether or not the sum of the contribution rates (α + β) is 0.2 or more (step S6). If not, the question items are too many and the first principal component A,
Since it is determined that the contribution rate of the second principal component B has been reduced, the question items that are not directly related to the advertisement are deleted from the survey items (step S7), and the process returns to step S4 and returns to the principal component again. Analysis is performed to obtain a first principal component A, a second principal component B, and contribution rates α and β. And the sum of the contribution rates (α +
If β) is 0.2 or more, the coefficients k ₁₁ , k ₁₂ , k ₁₃ ... of the eigenvectors of the first principal component obtained by the principal component analysis are graphically displayed on the display device as shown in FIG. . Alternatively, this graph is created by the printer 18 (step S8).

Further, the coefficients k ₂₁ , k ₂₂ , k ₂₃ ... Of the eigenvectors of the second principal component are displayed as a graph on a display device as shown in FIG.

The scores of the first principal component and the second principal component of each sample are obtained by substituting the scores for each survey item of each sample into the third and fourth formulas, as shown in FIG. ,
On an orthogonal coordinate system in which one of the first principal component and the second principal component is the vertical axis and the other is the horizontal axis (in the example of FIG. 7, the vertical axis is the first principal component and the horizontal axis is the second principal component) , A purchase intent / CM favorable plan distribution map, in which the principal component scores of each sample are plotted, is created, displayed and printed out (step S9).

In the graph of the eigenvector of the first principal component shown in FIG. 5, as a questionnaire item was designed in advance,
The “motivation for purchase” and the factors constituting this principal component appear in a negative or positive direction with a considerably large absolute value. This clearly shows that the eigenvector indicates the willingness to purchase the product as designed. A plurality of survey items having a coefficient with a large absolute value, opposite to the sign of “purchase motivation” in the first principal component, are items that need improvement in order to improve the motivation to purchase. Therefore, if you read a survey item with a large absolute value of the coefficient of this unique vector,
For example, it can be understood that a person who does not like a historical drama as a television program has a high desire to purchase the product. Further, it can be seen that the characters are unsuitable, and the color of the screen and the scene settings are working in a direction to reduce the purchase will. Also, looking at the coefficient in the positive direction of the opposite sign, for example, it is found that the screen switching of the advertisement is too early and requires improvement,
It is understood that narration of M and advertisements needs to be improved.

As described above, from the graph of the eigenvectors of the first principal component, the item that increases the willingness to purchase and the item that decreases the willingness to purchase the product have no relation to the willingness to purchase (the absolute value of the coefficient becomes 0). (Close) items are found as specific numerical values.

In the graph of the eigenvectors of the second principal component shown in FIG. 6, as previously designed, the "CM favorable" and the components of this principal component appear in a negative or positive direction with a considerably large absolute value. I have. It is clear that this eigenvector indicates the CM favorability for the product as designed. Also in this case, a plurality of survey items having a coefficient having a large absolute value and a sign opposite to that of “CM liking” in the second principal component are items that need to be improved in order to improve CM liking. When you load a survey item with a large absolute value of the coefficient,
For example, it can be seen that the sound effect of the advertisement, the favorable setting of the screen setting, and the non-boring of the CM contribute to the favorable CM. Also, looking at the coefficient in the negative direction of the opposite sign, for example, the advertisement is good for men but poor for women, the screen is switched too quickly, the character needs improvement,
Specific improvement measures for the CM, such as reducing the number of appearances of the product, are found.

As described above, there is no relation between the item for increasing the favorable impression of the advertisement from the eigenvector of the second principal component, the item for decreasing the favorable impression, and the CM favorable impression (the absolute value of the coefficient is close to 0). Items have been found as specific numerical values.

Further, the graph of FIG. 7 shows the relationship between the willingness to purchase each sample and the CM liking, indicating that a person who has a good liking for the commercial increases the willingness to purchase the advertisement product. Conversely, this indicates that a person who did not like this commercial would reduce their willingness to purchase the advertisement product.

In the above-described example, in order to increase the likelihood of the advertisement and increase the willingness to purchase, in addition to decreasing the tempo of screen switching in this TV advertisement, selecting a program to be provided, changing a character, and merchandise being commercialized You can get improvement measures such as reducing the number of appearances.

In the above-described example, an example of diagnosis for a television advertisement has been described. However, the present invention relates to, for example, a flyer advertisement, a newspaper advertisement, a magazine advertisement, a website itself, an advertisement on a website, a street advertisement. Ads, radio ads, catalogs, leaflets, brochures, free samples,
Demonstration advertisements, advertisements by notices, advertisements related to books, advertisements by mail, advertisements by e-mail, advertisements in cars, advertisements by signboards, broadcast advertisements, advertisements by mobile phones, display advertisements by containers and packaging, display advertisements by packages , Poster ads, logos, body ads,
Advertising by airborne objects, vehicles, structures, cloth,
The present invention can be applied to various advertisements such as an electric bulletin board advertisement, a video advertisement, an advertisement based on a handout, an advertisement based on a gift, and an advertisement based on a telephone.

The present invention can be applied not only to advertisements for products and services, but also to corporate advertisements and the like.

[0039]

According to the present invention, the effectiveness and effect of advertisements, such as the improvement of willingness to purchase and CM liking, are determined by diagnosing advertisements to be targeted through principal component analysis from the responses of randomly selected customers and consumers. It is possible to specifically determine a factor for improving the advertisement and facilitate the improvement of the advertisement. Since the effectiveness of the advertisement is determined by directly taking in the opinions and impressions of customers and consumers, it is possible to improve the advertisement more effectively.

[Brief description of the drawings]

FIG. 1 is an example of a computer system that implements an embodiment of the present invention.

FIG. 2 is a processing flowchart in the embodiment.

FIG. 3 is a diagram illustrating an example of an input data table.

FIG. 4 is a diagram illustrating an example of a correlation matrix.

FIG. 5 is a diagram illustrating a graph of an eigenvector of a first principal component in the example of the embodiment.

FIG. 6 is a diagram illustrating a graph of an eigenvector of a second principal component in the example of the embodiment.

FIG. 7 is a diagram showing a principal component score of each sample in the example of the embodiment.

[Explanation of symbols]

 10 Computer 11 Processor (CPU)

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Hiroto Sato 105-1, Yabashi-cho, Kusatsu-shi, Shiga Prefecture (72) Inventor Takaharu Fuchi 648, Noji-cho, Kusatsu-shi, Shiga Prefecture

Claims (4)

[Claims] 1. A questionnaire survey in which an item that is a factor in diagnosing and evaluating an advertisement is performed after a target person who is a target of the advertisement watches the advertisement,
The result is input to a computer and stored in a memory, and the main component analysis is performed by the computer based on the data stored in the memory, and either one or both of a purchase intention or a CM liking is obtained as a main component. An advertisement diagnosis method for diagnosing an advertisement based on the obtained data of the main component. 2. The method according to claim 2, wherein the questionnaire item is purchase intent or C
2. The advertisement diagnosis method according to claim 1, wherein the questionnaire is designed such that one or both of the M favorable sensitivities is a component of the main component. 3. The items of the questionnaire are designed such that one of the purchase motivation and the CM liking is the first principal component and the other is the second principal component, and the first and the first items obtained by the principal component analysis are designed. 3. The advertisement diagnosis method according to claim 2, wherein an eigenvector of the second principal component is obtained, and an advertisement is diagnosed based on the eigenvector. 4. A plot of the principal component scores of each sample on a coordinate system having one of the first and second principal components as a horizontal axis and the other as a vertical axis, and displays or prints the plotted values. 4. The advertisement diagnosis method according to claim 3, wherein the advertisement is diagnosed by referring to the advertisement.