Factors determining consumers’ perception of advertising messages in social networks

Размещено на http://www.allbest.ru/

Размещено на http://www.allbest.ru/

National Research University Higher School of Economics, St. Petersburg

Factors Determining Consumers' Perception of Advertising Messages in Social Networks

Dzhuletta Ovanesian

Abstract

Advertising messages in online marketing are an important part of marketing communications. One of the most growing Internet segments nowadays is social networks [Zdanovskay, 2012]. In Russia, according to the VTSIOM (2018) statistics, the most widespread social network is Vk.com. Plenty of research articles covered information on the cases of international social media like Facebook and Twitter. Therefore, our research will be conducted in Vk.com. The goal is to reveal the factors determining clients' perception of advertising messages. The research method used in data collection is mixed content analysis as it allows us to analyse advertising messages by our key characteristics - media type and content. The perception was measured by 15 metrics, among which are: CPM, CTR, eCPC, likes, reposts etc.

After the conducted correlation analysis in IBM SPSS , we determined that among the factors influencing consumers' perception of advertising messages are as follows: presence of picture, presence of photo, media type, presence of nature, presence of man, presence of woman, number of signs, number of words, nausea and learning material.

Key words: marketing communications, online marketing, social networks, Vk.com, advertising messages, consumers' perception, media type, content.

Content

• Introduction
• 1. Literature Review
• 1.1 Social Networks as the Part of Marketing Communications
• 1.2 The Creation of Advertising Messages
• 1.3 Clients' Perception Measurement in the Context of Social Media
• 2. Research on Determination of Factors Influencing Consumer's Perception of Advertising Messages
• 2.1 Research Design
• 2.2 Research Methodology
• 2.2 Data Collection
• 3. Description of Received Results. Determination of Factors Influencing Consumer's Perception of Advertising Messages
• 3.1 Research Results
• Conclusion and Discussion
• References
• Appendixes

Introduction

• Advertising messages in online marketing are an important part of marketing communications. In Russia, according to the ACAR (Association of Communication Agencies of Russia), the budget spent on online advertisement accounted for 35 billiard of rubles in 2017, which is 23% higher than in the same quarter of 2016. [Kozyrev, 2017] To make these investments be paid off, marketers should pay enough attention to advertising policy in the Internet and, in particular, to advertising messages, because messages do not simply inform but affect the perception of current and potential consumers. Therefore, in this paper we determine factors which influence clients' perception of advertising messages.
• We focus on social networks advertisement because, according to the research of Zdanovskaya (2012), social networks are one of the fast growing sectors of the Internet. 74, 5% of users in Russia visit at least one social network during a month. Moreover, according to Microsoft research, approximately 77% of people born between 1979 and 1996 have its own page in social networks (Facebook, MySpace etc) and 28% of them publish posts. This population nowadays is about 21-34 years old and is able to make purchases which prove that marketing communications in the Internet can aim at this group of people. [Barry, 2009]
• Factors determining the process of spread of advertising messages have already been examined in the frames of several approaches by scholars [Kwock & Yu, 2013; Asur et al., 2012; Smith et al., 2012]. However, these research articles covered information on the cases of international social media like Facebook and Twitter. Moreover, every social network has its own features and peculiarities of its messages. Kwock & Yu (2013) have revealed that advertising messages' popularity do not depend on their content but only on media type. However, Kirakosova (2014) in her research paper determined that media type does not influence users' reaction. Nowadays to attract people's attention in a mass media and to bring important information to the audience users should read a message and then make some conclusion about a community. Ryzhkova (2010) in her research paper claimed that the numbers of symbols do not matter, but if the text and its content are interesting, people will read a message. Therefore, we suppose that the role of content was undervalued. Finally, the majority of research on this topic was conducted on the Facebook's data. The number of the audience of the most popular social networks in Russia, according to TNS, in 2013 has increased by 10% in relation to Vk.com, by 6% - to Ok.ru and by 4, 5 times for Twitter.[TNS Web Index, 2013] Therefore, it is interesting to study consumers' perception in Vk.com. Moreover, we have received an order for research of contextual advertising messages for the online business education company, the main market and the main channel of communication of which is Vk.com.
• There are several research articles devoted to this topic in Russia, but they mostly cover the connection between messages and their authors, observe non-advertising types of messages [Kirakosova,2014] or brand associations [Anishenko,2015]. There is no scarcity of research activity in Russia concerning advertising messages and its interconnection with consumers' perception in social network Vk.com. The results of this paper will be useful for marketers, SMM - managers, copyrighters etc. As companies have to adapt to marketing trends, they shift online marketing to achieve their marketing goals. It is important to understand the factors that determine consumers' perception of advertising messages and to formulate them clearly.
• The units of study are advertising messages in social network Vk.com.
• Advertising texts are significant at the various stages of the buying process. The goal of the paper is to reveal the determinants of clients' perception of advertising messages. The research question is the following: «What factors determine the consumers' perception of advertising messages?»

To reach the aforementioned aim, the following objectives have to be attained:

1. to determine the role of social networks as a part of marketing communications;

2. to analyse and compare models of advertising messages' content analysis;

3. to depict possible ways to measure consumers' perception of advertising messages;

4. to develop the methodology for the content analysis of advertising messages of Vk.com;

5. to receive outputs and justify them.

The goal will be achieved with the help of the mixed content analysis of the advertising messages of online business education company in Vk.com. The data will be statistically processed through correlation analysis in IBM SPSS.

social network advertising message

1. Literature Review

1.1 Social Networks as the Part of Marketing Communications

Marketing communications are an essential part of marketing policy. However, during the last decades they have considerably changed. First of all, Integrated Marketing Communications (IMC), the communications consisting of the combination of several marketing instruments, have recently appeared. Numerous scholars agree that IMC are an indispensible feature of a good marketing [Pelsmacker et al., 2004; Prasad & Sethi, 2009]. Pelsmacker et al., (2004) highlight two effects of IMC: interactivity as a possibility of audience to reply to communications and synergy effect. Prasad & Sethi (2009) and Pelsmacker et al. (2004) underline that instruments must not be in conflict but have the same direction.

There is no consensus among scholars regarding the classification of MC. For example, marketing communications were classically divided into BTL (or below the line) and ATL (or above the line). Nowadays the third type appeared known as TTL (or through the line) which blurred the line between the previous two types: several marketing channels are combined and not divided into BTL and ATL. [Wells et al., 2009]

Pelsmacker et al. (2004) divide communications into advertising, sales promotions, sponsorship, public relations, point-of-purchase communications, exhibitions and trade fairs, direct marketing communications, personal selling and the Internet. This classification is not admitted accurate by us because nowadays the Internet has become a part of other communication tools. That is why it is also important to address the issue of Internet communications because the spread of the Internet and smart phones have greatly changed traditional marketing communications. The main feature of the Internet is its many-to-many communications' character - if earlier consumers received advertisement without a possibility to answer on it, now consumers can interact with a seller and with each other. [Pelsmacker et al., 2004].

Nowadays Social Media Marketing (SMM) - the promotion of a product through the social Internet platforms (social networks, blogs, micro blogs etc) is determined as a separate kind of marketing communications. [Solomon & Tuten, 2012] The process of spread of information in the frames of social media is crucial for various fields. At first - for marketing field because posts about company and its products, spread in the Internet, influence its brand image.

The various classifications of social media exist nowadays. One of them was created by Curtis Foreman, Hootsuite's content editor, in 2017 and divides social media into 10 types. They were fulfilled by Russian social media as well. They are reflected in the Table 1.

Table 1 Classification of Social Media types by Curtis Foreman

Type of social media	Function	Example
Social networks	Communicate with people	Facebook, Twitter, LinkedIn, Vk.com, Ok.ru
Media sharing networks	Share media	Instagram, Snapchat, YouTube
Discussion forums	Discuss ideas, news	Reddit, Quora, Digg
Bookmarking and content curation networks	Open, save, share actual content	Pinterest, Flipboard
Consumer review networks	Write or read reviews on different companies, services, products	Yelp, Zomato, TripAdvisor
Blogging and publishing networks	Writing and publishing the content	WordPress, Tumblr, Medium
Interest-based networks	Share with each other interests , news	Polyvore, Etsy, Fancy
Social shopping networks	Purchase online	Goodreads, Houzz, Last.fm
Sharing economy networks	To offer or use services or products	Airbnb, Uber, Taskrabbit
Anonymous social networks	Communication without revealing the personality	Whisper, Ask.fm, After School, Sprashivay.ru

*based on: Foreman, C. (2017). 10 types of social media and how each can benefit your business. Hootsuite.

The social media types which classically appear in news are social networks (Facebook, Vk.com), blogs (Blogger, LiveJournal), micro blogging (Twitter). In relation to the last one, sometimes Twitter is related to the social networks, sometimes - to micro blogging. In the frames of this paper it is not crucial; moreover, nowadays the technologies are developing, the functionality of various platforms increases that is why the lines between types are blurring as it was noted by Foreman (2017).

As it is claimed, social networks are not the oldest type of social media but they determine social media nowadays [Foreman, 2017]. Wasserman et al., (1994) define social network as a social structure consisting of various social subjects (people and organizations), connections and interactivity between actors. Neyaskin (2010) defines social networks as a category of websites giving a possibility to publish exchange and discuss content by a wide range of users. Oxford Dictionary (2018) defines social networks as special website or application which allows users to communicate with each other and exchange information by publishing information, comments, messages, pictures etc.

According to the research of analytics of IBM, even in 2011 more than 50% Internet users in Europe (from 16 to 64 years old) used social networks while deciding to make a purchase. More than the third of users were involved into the discussions and comments on the information about the goods and services in social networks. The number of users of social media is increasing because the field is constantly developing. Social networks are the strong marketing instrument not only in Europe but also in Russia. [Forge & Steegenga, 2011]

According to the research of Zdanovskaya (2012), social networks are one of the fast growing sectors of the Internet. 74, 5% of users in Russia visit at least one social network during a month. According to the All-Russian Center of Studying Public Opinion (VTSIOM) survey results, opinions of people on goods and serviced cause a trust of 62% of Russian Internet users, while information in online shops is trusted in the percentage of 32, TV advertisement is being trusted in 31% and Internet advertisement - only by 18%. [VTSIOM, 2012]

Then, VTSIOM (2018) presented research data on the use of social networks by Russians in 2017. It turned out that 45% of respondents over 18 years of age use at least one of the social networks almost every day, 62% - at least once a week. The highest level of involvement was expected among young people: 91% of people from the group of 18-24 years old use social networks almost every day; among people of 25-34 years old - 69%, in the group over 60 years - only 15% use social networks every day. The most widespread coverage in Russia has Vk.com network - 28% said that attend it daily, the second place is occupied by Odnoklassniki (19%), the third - by Instagram (14%). Vk.com and Instagram were the most «youth» venues. The largest share of the audience Vk.com users - 40% - accounted for by people aged 25-34 years. Among Instagram users, 38% are aged 18-24 years old, 37% are aged 25-34 years old. Over the six months, the number of users preferred Vk.com has grown - in august of 2017 60% of users spent their time there, whereas this number reached 65%, as Levada Centre noticed. [Shmyrova, 2018]

The popularity of different social media varies from country to country. For example, Facebook, Instagram and Twitter are famous worldwide, whereas Vk.com and Ok.ru are well known in our country. According to the data of enormous Russian services of Internet Statistics (Yandex.Metric, SpyLog/Openstat, LiveInternet, Hotlog, Rate of @Mail.ru and SEO-AUDITOR), the most popular social network in Russia in the period from 03.2017 till 02.2018 was Vk.com. The next social networks relatively are Facebook, Odnoklassniki, Twitter and YouTube. Vk.com reached its highlight at the February of 2018 when the popularity was at the amount of almost 63% [Rating of social networks, 2018].

The data described above were taken into consideration; therefore, the analysis will be based on advertising messages in Vk.com - the most popular and growing social network in Russia.

To summarize, marketing communications have faced various changes. Integrated Marketing Communications (IMC) has appeared as well as new marketing tools; consequently, some classifications have become irrelevant. Social media as an actual way to promote products through SMM which is determined as a unite type of marketing communications. Significant part of social media is social networks - website allowing users to communicate with each other by posting various kinds of content. [Oxford dictionary, 2018; Neyaskin, 2010] In Russia the most popular and important social network is Vk.com. Therefore, it has been chosen as a base for future analysis.

1.2 The Creation of Advertising Messages

In this paragraph the role of advertising messages in advertising policy will be examined. Therefore, some crucial points on the creation of social network posts and various models of classifications of advertising messages will be revealed. In conclusion the best classifications of advertising messages and their metrics will be formulated.

Key element of advertising communication is an advertising message. It plays an important role due to its influence on a customer. It is determined as a vital element of achieving advertising goals. Wells et al. (2009) claim that advertising message is aimed to represent a communicator to the target audience. It is a carrier of different character influence which has a particular form (textual, visual and symbolic). The authors define it as an element of advertising communication being a carrier of informational and emotional content, brought by the process of advertising communication.

Some authors [Grшnhaug, Kvitastein & Grшnmo, 1991; Tellis, 1998] claim that all general theories of advertising measurements come to the point of hierarchy-of-effects model where the variables are included in a special order. The hierarchy-of-effects models by Scholten, Lavidge & Steiner, Smith & Swinyard, Bergkvist & Melander, Rossiter & Percy have been examined. The DAGMAR model by Scholten (1996) includes such elements as awareness, comprehension, conviction and action. Another one by Lavidge & Steiner (1961) consists of awareness, liking, preference, conviction and purchase. The alternative models were suggested by Smith & Swinyard (1982) and consist of the similar variables but in other order: cognition, affect and conation. «The six step effects sequence» model (four steps reflect consumer's attitude to the product and two are connected with the market and profit) by Rossiter & Percy (1997) has been used in the research by Bergkvist & Melander (2000).

However, Smith & Swinyard (1982) stress that the hierarchy-of-effects models were argued because they do not take into consideration a market situation. That is why the order of the options should not stay constant for different markets, brands and situations. Despite of it, hierarchy-of-effects can be applied to identify key variables [Bergkvist & Melander, 2000].

Downing (2000) reveals that some researchers compare advertising with literacy writing because both create fictional worlds to build communications. Another scholar [Marrone, 2001] marks that language of advertisement depends on the everyday language because it uses idioms and syntactic rules. According to Cook (1992), advertising messages as the part of advertisement are rather concentrated on linking the products with effects, people and desirable properties than on real characteristics of a product.

Two formats of advertising communications from the content point of view are revealed by Downing (2000): textual and visual. It is noticeable that only at the end of 1990-s visual elements of advertisement were admitted as important as textual. Moreover, Bianchi (2011) marks that image is more complex than text. It has two forms: denotative and connotative; in other words, symbolic or literal. Sometimes textual part helps to understand visual element; often they are used together in an advertising message. In our research these elements and their content will be examined in details.

Textual advertising messages

With respect to textual messages, researchers mention the use of the linguistic analysis like graphology, phonology, lexis, syntax and textual rhetoric, puns, metaphors and metonyms [Downing, 2000]. Pardillos (1995) describes three elements of the text creation: the use of pronouns, ellipsis and presupposition. For example, personal pronouns have an effect of involving reader into a situation as an advertisement is addressed directly to him. The «you» in advertisement refers both to characters described in a fictional situation and to addressees, the "I" is related to manufacturer, companion or an expert, and “he" or "she" is an allegory of competitor. Cook (1992) stated that the use of them depends on context. Other authors [Downing, 2000; Semino, 1997] also mention context related to advertising texts. It can be defined as a relationship between participants, which is determined by social factors. It means that advertisement should be understood by its target audience and suitable communication manner should be applied.

If we go deeper into textual advertising messages, Ryzhkova (2010) characterized them by the following features: the number of signs, the number of words, the absence or presence of paragraphs, the size of the font, the punctuation, the presence of comparative turns and other literary methods, the presence of interrogative constructions, the repetition of the main idea, the presence of keywords necessary for SEO optimization, the emotionality of the text.

In the research Ryzhkova (2010) claims that the number of signs and words do not matter, but the type of expressiveness, title, paragraph logic and punctuation do. Gordon & Gordon (2012) mark that the text is always influenced by used fonts:

ѕ Ordinary - tells;

ѕ Italics - explains;

ѕ Bold - prescribes;

ѕ Fat - reports the most important idea.

According to Suchkova (2016) verbal part of a text consists of a motto, headline, core advertising text and repetition of the main idea. She claims that text should be simple and laconic. If the length is long, message is divided into paragraphs. Headline and the reduming phrase should be separated from the main body by line spacing. The author also protects the thought [Vasilevich & Ledeneva, 2005] that the general perception of advertising message depends on a visual presentation of a text.

Copyrighters use special metrics for evaluating various texts. [Anokhin, 2017]

Originality - is a core variable; it let us check whether a text was written by copyrighters themselves or it was copied from other sources. It is important to understand that it is practically impossible to achieve 100% of originality because of the use of standard phrases and quotes. In this case we can orient on the number of 80-90. There are several instruments for measuring it as follows: Content watch, Antiplagiat, Findcopy, Text.ru, Advego plagiatus, Etxt. Text.ru was remarked because it does not use the standard algorithm for checking the uniqueness of text by the shingles method (which is used by the most services), what let the service to successfully determine a substandard rewrite.

Nausea - academic nausea is the ratio of the number of most often used in the document words' repetitions to the number of words in the entire text. If a rate is large (more than 5-7%), then text is a re-spam.

Number of unique words - the number of words which are used once. Large texts contain more unique words which let them bring more traffic.

Stop-words («water» content) - are words and phrases which can be removed from text without missing any sense. For instance, «it is not difficult to guess», «without any doubts». Sometimes they can be used but if a text is commercial or informative, they do not have any weight. One of the free services which can check «water» content is Сontentmonster. Other popular service is Glvrd which shows the relation of stop-words to all words. However, it evaluates not text quality but only the ratio. It was invented as a service for those who work with informative style. According to Shardakov (2016), a customer should designate the stop-words rate (if the text is checked in Glavred) not less than 7, because if the rate is near 10, it increases risk to obtain «dry» and «lifeless» material. [Shardakov, 2016]

In relation to our research, characteristics connected with text originality will not be used because we conduct content analysis of real published advertising messages, therefore, the uniqueness percent would be small.

Thus, the metrics for evaluating textual part were formulated, according to the various researches [Ryzhkova, 2010; Gordon & Gordon, 2012; Suchkova, 2016; Vasilevich & Ledeneva, 2005; Shardakov, 2016; Anokhin, 2017]:

ѕ Number of signs

ѕ Number of words

ѕ Nausea

ѕ Water content

ѕ Presence of a title

Visual advertising messages

To make a closer look to the visual part, also several characteristics can be defined. In the research of Suchkova (2016) devoted to the perception of graphic advertising messages, she defines various graphic instruments: color, font, composition. Berman's (2010) also reveals the elements of design which are fuller. Apart from the mentioned above by Suchkova (2016) they include: format and mass media, form, line and contrast. In relation to colors, warm colors are contrasted to cold. As Knabe (2006) notices, cold colors shade off and create the effect of space depth, and warm colors approach brings objects closer. During laboratory experiment scholars collected data on the eyes' movements during viewing advertising messages in phone books. As a result, some messages were not noted by consumers: they noted approximately 93% of advertising messages. Colorful messages caught more attention and were examined by 21% more than black and white messages. [Lohse & Wu, 2001]

In our research several metrics for evaluating visual elements in messages will be connected with: colors, presence of corporate style (for eg. logotype, fonts) and presence of a human and nature. [Suchkova, 2016; Knabe, 2006; Lohse & Wu, 2001]

According to the content of messages, Asur et al. (2012) supposed that mass content will be more popular than content interesting for a small proportion of people. The main hypothesis was that messages' popularity depends on the content it includes. Based on the messages of social network Twitter, brought during 50 days, with the help of classification and regression analysis the authors studied to what extent the spread of a message depends on the following characteristics:

ѕ The author of a message (resource);

ѕ Content of a message (category);

ѕ Type of writing (subjective/objective);

ѕ Notifications about famous people.

The initial supposings that the content influences message's popularity and subjectivity of a language is significant for a message's spread were not confirmed. Content occurred not as important as a resource (author who posted a message). In other research conducted by Kwock & Yu (2013) authors used other approach. In the frames of content analysis they have analysed the popularity of messages of 12 restaurants in social network Facebook. They tried to determine the most typical words which are often used in successful and popular messages. For example, the key words for the most popular messages became the following:

ѕ Words describing the menu («sandwich», «lobster», «chocolate»);

ѕ Special offerings in restaurants and time characteristics («October», «Friday», «august»);

ѕ Questions to customers («how», «who», «try»).

The least popular messages included the words related to the marketing campaigns («victory», «winner», «chance» and «contest»). From the one side, results helped to conclude that marketing messages are less popular. From the other side, this approach to the analysis leads to the following problem - these lexical groups are very difficult and practically impossible to find, they are dependent on the topic of messages. Furthermore, the same words can be used in various meanings and in messages of different topics in popular as well as unpopular communities. Kirakosova (2014) in her paper also agrees with that conclusion about applying the approach of content analysis of particular words. It refers us to the problem of context's understanding which was raised by Cook (1992), Downing (2000) and Semino (1997). It proves that even through the decades this issue remained. [Kwock & Yu, 2013]

However, despite of the similar relationship to the use of particular words, the question about the content of the messages from the position of topics, visual and textual characteristics and the goals of the posts is still open. Therefore, the classifications of posts according to the main information in them will be revealed in the next part of the paragraph.

The research of Millward Brown and Dynamic Logic have revealed that the most expected users' variables of the brand in social media are traditional postings, reliable news about the brand, information on new products, competitions and special offerings. It is remarked that entertainment, diversity, innovations, interactivity and the feeling of connection highlight the social group and attach a user. [Specialists WFA, Millward Brown, & Dynamic Logyc, 2011]

Purohit et al. (2015) in their research «Intent Classification of Short-Text on Social Media» have mentioned the classification of the content as navigational, informational and transactional. In their research they have divided posts into the following measures: seeking, offering, none (neither seeking nor offering). The pattern of the design was based on semantic-syntactic approach to identify the intentions of the users. For instance, the verb «have» with a noun means to be an offer. When «do» is mentioned with «you», the authors notice the seeking expression. The following gradations were used while evaluating the tweets: {Request to get, Offer to give, Both request and offer, Report of past donations, None of the above, Cannot judge}. [Purohit et al., 2015]

However, as it was remarked above, context is a thing of upper importance when creating a message. We cannot rely only at the verbs used because the structure of the text and words' use vary. What is important for us is that a message can include, for example, both requesting and offering meanings. Finally, sometimes the meaning cannot be judged.

Ideally, advertising text consists of an advertising message (what the text is about), advertising treatment (to whom it addressed and how the addressee affects the message's style) and call to action (what should be understood by a recipient). Therefore, advertising message implements 3 functions: informative, communicative and suggestive. [Uhova, 2012]

Anishenko in her paper determined 12 metrics which characterize the means through which the companies communicate with the audience.

They are the following:

a) Publication of news about brand/company - the most popular instrument for communication with customers and informing them. Often a message is combined with a picture to catch users' attention.

b) News about discounts and sales - the traditional methods of promotion of sales, the main aim of which is to attract users' attentions to a product. 20% of customers read this kind of information.

c) Posts about competitions which can be classified as an instrument of promotion of sales and attracting of attention. Through such instruments as «like» and «repost» a social network can increase the number of messages readers and visitors of a community and enhance an involvement.

These three metrics are considered to be core because according to the analysis of communities in social network Vk.com by Anishenko, brand communities are created for bringing the information to their customers, for increase of customers' involvement and the enlargement of the audience. Besides the mentioned means, the companies use other instruments to be noticed among the competitors.

They are the following:

d) Entertaining content (humor, funny pictures, photos, videos). Using a brand in an entertaining content, a company achieves user's attention even if a customer saw a post in news but did not visit a community. At the same moment a company is in risk because too much of entertaining information can harm the image of an organization. That is why companies which are not associated with an entertainment publish amusing content but without a connection with a brand.

e) Presence of photo and video albums characterizes the fullness of a community's content. It enlarges the time which a costumer spends on a brand.

f) News which is not connected with a brand. According to the research of Pew Research Center, 1/3 of adults in the USA get news from Facebook. Furthermore, FOM has determined that social networks in Russia are included in «Top 10» of the most popular news resources among the youth. Often companies publish news which is close to a company's field and which widen the audience's horizons and attracts user's attention. [Zhebit, 2011; Mereminskaya, 2013]

g) Useful information which is often combined with information about a company.

h) The community's wall which is open or close for users' publications. This variable characterizes the openness of a company for communications with its consumers. At the same time the risk of publishing advertising or obscene posts occurs. Moreover, the content of a company itself can be lost and moved to the background. Therefore, it is rarely used, often only in small communities with a highly loyal audience, without «casual» visitors.

i) The comments to the posts. If they are open, users have a possibility to express their opinion which increases their involvement. Moreover, it allows people to communicate with each other - it helps the main goal of social networks (to communicate) to be achieved. Finally, a company gets a feedback in the form of comments which can be interpreted through the content analysis.

j) The presence or absence of discussions - the method of communication between company and users as well as users with each other. It increases the frequency of community's views and feedback collection. However, the discussions should be controlled and the questions should be answered in time. Therefore, there is a need in a specialist who will be responsible for it.

k) Communication of a company's representative with users. It helps to answer consumers' questions, to work with complaints and to increase loyalty by creating a friendly atmosphere. A specialist should have full information, be polite and correct, as he speaks on behalf of the company and is responsible for its image.

The variables were united into 3 groups:

ѕ Information aimed at the promotion of a brand, advertising information (news about company/brand , new products; information about sales and discounts; competitions; entertaining content linked with a company);

ѕ Entertainment and informing (entertaining content, which is not connected with a company, news on common topics; useful information :photos and videos not about a brand);

ѕ Communication with users, possibility to express the opinion (open wall, open comments, discussions, communication between users and company's representatives).

According to the finding's result, Anishenko (2015) on the case of 212 brands and organizations has found after the consumers' survey that the most interesting company's activities for users from higher to lowest are respectively:

1. Brand news;

2. Discounts, sales and contests;

3. Photos, connected with a brand;

4. Humor, funny pictures;

5. Useful information (not about a brand);

6. News on general topics;

7. Surveys.

At first, users would like to see information about brand news. Then, they are interested in entertaining content and useful information - otherwise, a communication with a group becomes boring or useless. Additionally, users noted that they would like to see the following information: company's history, information on sales and new products, news about a company and news about similar products. They liked content with videos and audio files less. It can be caused by a time laborious character of this content. The result is similar to the one Kwock & Yu (2013) received. The research has also shown some conclusions about the activism of the audience: 12 % of users are not active at all, the others like posts but the majorities do not communicate with others.

Other type of a content which can be defined is user generated content or UCG - the material created by users of a community themselves. Smith et al. (2012) have devoted their research to this type of content in Facebook and Youtube and also used a content analysis as a research method.

In the frames of this paper we conduct a research on the content of advertising messages and their nature; we do not focus on discussions and multimedia channels of the community (like albums with photos and videos). We classify the advertising messages according to their content and study in details the content of written and graphic messages because this aspect is not revealed in Russia and, according to this paper's author, is undervalued.

In general, the goal of advertising is persuasive, i.e. to urge people buy something. However, Cook (1992) argues that advertisement also deliver amuse, information or warning. The emotive function also plays a significant role. In this way, the classification of advertising messages can be formulated based on their main goal of the message. The similar approach to it was used by Anishenko (2015) as she classified the company's actions and stated the aims of these actions, and by Uhova (2012).

The model used by Anishenko (2015) will be used in the following research as it is:

ѕ Detailed;

ѕ Simple;

ѕ Suitable and was used for Vk.com.

However, it will be corrected and complemented to fully meet the aim of this research. Thus, on the base of previous actual research of Anishenko (2015), Smith et al. (2012), Specialists WFA, Millward Brown, & Dynamic Logyc (2011), we can divide the posts into several groups. They are the following:

ѕ Promotional posts - the main goal is to promote something (description of products and services, posts about payments of products, discounts, sales, posts about usage of products);

ѕ Competitions* - can be define as a part of promotional content but due to its spread it can be classified as a unite type - the main goal is to widen an audience reach and increase its involvement (contests, quizzes);

ѕ Entertaining posts - the main goal is to remove intellectual tension and cheer up subscribers (humor, anecdotes, quotes, stories);

ѕ Learning material - the main goal is to educate the audience and increase its loyalty (learning articles, the overview of learning resources and services, webinars, description of a company's experience, inscriptions and myths);

ѕ News material - the main goal is to attract users' attentions to a product and to inform them (news about company's field, news about company itself and its products, vacancies);

ѕ UGC (user generated content) - the main goal is to humanize a brand (reviews of products, users' questions published on a company's wall);

ѕ Interactive (engaging content) - the main goal is to involve an audience and increase their activism, create the atmosphere of friendly and live communication (daily posts, surveys, night or daytime chats, request for advice or giving of advice, discussions under a post, games, provocations, complaints).

The model of Uhova (2012), where the messages were divided into: informative, communicative and suggestive will be applied because of its laconic character which reflects the core advertiser's goals. They are formulated as 3 metrics: advertising message, advertising treatment and call to action.

The next category which was studied by several authors [Kwock & Yu, 2013; Kirakosova, 2014; Asur et al., 2012; Uhova, 2012; Cook, 1996; Downing, 2000] was media type of the message. Nowadays in the era of technologies and rapid development of social networks classifications of advertising messages are broader than textual and visual. Therefore, other works should be examined. In the next part of the paragraph they will be revealed.

Kwock & Yu (2013) in their paper have considered characteristic of a message as media type determined as a post's format: text, text with a link, post with a photo and post with a video. They claim that it is important for post effectiveness whereas the content is not. Their analysis has shown that textual posts and posts with a photo get more responses from Facebook users than posts with links or videos. It can be caused by 2 reasons: first is that users are more attracted by simple and common messages with text and photo rather than those which require more time for studying it (website or video). The second possible cause is that after watching a video or serving a website users forget to like a post or write a comment. This research proves that post format is significant for users.

Kirakosova (2014) in her research devoted to successful promotion of a message in social media used the following model of a messages' structure (media type), similar to the Kwok's model:

a) A presence of a link to other Internet source;

b) A presence of a picture;

c) A presence of a video.

Despite of her assumption about the role of media type in message's popularity, the coefficient of determinacy has shown that the list of independent variables explains messages' popularity only by 0, 1% which means that popularity (measured by likes) is not statistically significant depend on a media type of post. Furthermore, the message spread (measured by reposts) is not determined by media type. Moreover, in general, a textual message is able to get 3, 4 likes but the presence of additional elements like videos or pictures does not guarantee the higher level of popularity.

Nowadays this division can be considered as important. Media type was admitted valid for comparing messages but the results of authors were different. Therefore, it represents an interest - to check whether the perception of messages is affected by its media type on the example of the Russian social network.

All in all, 2 concepts for evaluating advertising messages have been chosen - they are the following: media type and content. The messages will be differentiated by the following parts: textual, visual, video, audio and link to a website. Visual elements divided into photo, picture (art, drawing), gif, info graphics and meme.

Content characterizes textual and visual parts of messages and content classifications (type) according to the topics. Textual and visual parts will be studied in details because the other formats require huge time resources and even special technical equipment. Textual metrics are the following: number of signs, number of words, nausea, water content and presence of a title. Visual metrics will contain information on: colors (black and white or colourful; warm or cold), presence of corporate style (for eg. logotype, fonts) and presence of a human and nature. Content type was also divided into 2 groups - according to their topic and content: promotional, competitions, entertaining, learning, news, ugc, interactive; and according to the structure of advertising message: advertising message, advertising treatment, call to action. The term of perception and its measurement will be revealed in the next paragraph.

1.3 Clients' Perception Measurement in the Context of Social Media

In the following paragraph the term of perception, its connection with effectiveness will be examined. Moreover, the methods used for evaluating perception will be revealed as well as the methods of measuring effectiveness. As a result of the paragraph the method and metrics for evaluating perception will be defined.

Perception, defined as the «organization, identification, and interpretation ofsensoryinformation» in order to produce and grasp the given information or the environment [Schacter et al., 2011] and the understanding of a text by a recipient is a base for text's effectiveness. According to the researchers of advertising texts, Vasilevich & Ledeneva (2005), perception of a text can be considered as a first step of a complicated process the main goal of which is a client's decision to buy advertised product. Success of this step is directly connected with a text quality. It is worth noting that before reading an advertising message a person spends 1 second to understand the message's value. If a consumer decides to pay attention to this message, the next 3 seconds he is viewing an advertisement; starting from left high corner going down to the right corner and then to the center of a message. To sum up, it takes 4 seconds to perceive an advertisement. (Berman, 2010) Under a classical definition of perception lies a model of meeting expectations, according to which satisfaction appears when a consumer compares his expectations with a reality. In this case 3 states can occur:

ѕ Satisfaction (positive discrepancy);

ѕ Neutral relation (no discrepancies);

ѕ Dissatisfaction (negative discrepancy). [Bruner, 2009]

Then, it is common that advertisement with a human presence is understood better and in a more positive way because a person associates it with himself [Godfroi, 1996].

According to the levels of perception, Karl Yung (1991) classifies them into 4:

ѕ Logical;

ѕ Imaginary (emotional);

ѕ Sensory (perception organs);

ѕ Intuitive (imagination).

Suchkova (2016) in her paper reveals 2 levels: rational (logical) and emotional. With the help of modern technologies scholars can study brain's work. It became known that logic (left) hemisphere turns on later than emotional (right) hemisphere. Thereby, the starting point is our emotions which were caused by advertising messages. Logic is a continuation which searches for a reason to buy a product. [Lindsrom, 2010]

The clients' perception evaluation of advertising messages is the way of proving that advertising campaign is not conducted in vain. Result of a marketing campaign is toughly connected with a term of effectiveness, which is indirectly influenced by consumers' perception of advertisement. That is why the measurement of effectiveness will also be examined in the following paragraph.

Panin & Rebyazina (2016) in their research wrote about customers' satisfaction as a factor indirectly affecting effectiveness evaluation. According to the several researches, consumers' loyalty determines the level of other significant for a firm means [Chumpitaz & Paparoidamis, 2007; Flint et al., 2011; Lam et al., 2004]. In accordance with one group of empirical studies, satisfaction is positively correlated with the following unfinancial indexes: customer retention rate [Gustafsson et. al., 2005], share of repeated purchases [Anderson & Mary, 1993], performance of WOM [Szymanski & Henard, 2001] etc. Other researchers define that these factors influence financial indexes as well: profitability [Williams & Naumann, 2011], share of wallet [Cooil et. al., 2007] etc. It proves the connection between effectiveness and perception because satisfaction is a positive perception.

It is claimed that to evaluate advertising effectiveness correctly the line between a commercial success (increase of income and market share) and communicative effectiveness should clearly be stated. Commercial success is determined by a list of factors as goods and services quality, price, package, firm's reputation, competitors etc); only one of them is advertisement. It influences purchase making decision only not more than 1/8 of the whole factors' list. Therefore, direct link between advertisement and profit increase can hardly be established. Mentioned it all above, it is more correct to evaluate communicative effectiveness, not commercial. [Pirogova, 2000]

This point of view is admitted by the author of this paper because:

ѕ Consumer can like an advertisement but it will not surely lead him to purchase;

ѕ The future actions of a consumer on a purchase after watching an advertisement can hardly be measured.

Therefore, in this research we track a consumer on the stage of his perception.

Model of effective marketing communications suggested by Rossister & Persy (1997) as a part of hierarchy-of-effects models includes 6 stages of effectiveness achievement. They are the following:

ѕ User contact with a message;

ѕ Processing message information;

ѕ Effects of communication and positioning of the brand;

ѕ Actions of target audience;

ѕ Increase in sales or market share, creation of brand equity;

ѕ Making a profit.

The first 4 effects are called «stages of customer's reaction». [Rossiter & Percy, 1997]

The researched terms can be divided into the methods and the metrics used through the use of these methods. To start with, it was argued that the application of the traditional advertising models to the fundamentally new marketing platform is not applicable [Hoffman & Novak, 1996]. McQuarrie & Mick (1999) determine four historical approaches for evaluating advertisement:

a) the archival (the oldest method which usually brings a lot of advertising samples; studies conduct content analysis to determine the frequency of appearance of visual parts. It can also find the correlation between the specific elements and the audience's replies);

b) the experimental (it contains rigorous causal analysis with theoretical combination);

c) the reader response approach (this method is devoted to the information which consumers draw from advertisement; sometimes researchers conduct in-depth interviews);

d) the text-interpretive method (it is based on semiotic, rhetorical and literary theories. It studies visual and textual parts of advertisement as equal elements).

In relation to the methods of evaluating advertisement, the best method for our research was chosen content analysis as a common method used for analysis of advertising messages and also because it is possible to measure perception (communicative effectiveness), not commercial.

Content analysis was used in different fields. Hermeneutics and philology have been using it for many years for interpreting sacred and profane texts [Tiraldo, 2014; Krippendorff, 2004]. The political scientist Lasswell (1948) stated vital questions of content analysis: «Who says what, to whom, why, to what extent and with what effect?» Another figure, who made a contribution to the development of content analysis was Bernard Berelson. He defined the method as «a research technique for the objective, systematic and quantitative description of the manifest content of communication». [Berelson, 1952] Kolbe & Burnett (1991) describe it as a broadly admitted method for systematically measuring the content of communications which were recorded. Krippendorff (2004) defined content analysis as a research method for making reliable conclusions from texts (or other meaningful matter) to the contexts of their use.

...