Targeted Advertising on WeChat Official Accounts

This is a small research proposal about targeted advertising on WeChat official accounts (WeChat Channels).

1. Motivation

Users’ online activities become much easier to be collected by social media now. Based on this, the operators can deduce the users’ online preferences and behavior, which enables advertisers to promote their products to targeted customers precisely.

There are usually two different advertising channels on social media for advertisers to choose. The first channel is through social media’s official advertising platform. The social media provide some advertising space in users’ timeline for advertisers to buy. Facebook in the US and WeChat in China both have this kind of platform based on their big data analysis. The second channel is to post advertisements on those accounts with lots of followers, which make the advertisements appear in the followers’ timeline.

Since the operators of the social media have easier access to the users’ online activities data, the former channel seems more efficient to targeted advertising. However, the effectiveness of the later channel should not be neglected. Those users following the same account tend to have more similar interests and preferences. These homogeneous characteristics make the targeted advertising be possible. In the advertising market, there are many advertising agents are doing this work. They help advertisers to determine appropriate creatives and to find social media accounts followed by targeted consumers.

In China, more and more people are choosing WeChat official accounts as their major information source. Not only organizations and companies but also each individual can send articles through this platform to their followers once a day. Gradually, companies start to promote their products by posting advertisements on those accounts with huge number of followers. But whether these advisements are targeted to the right potential customers remains a question. Therefore, here we want to know the real effectiveness of the online advertising on this new social media platform and figure out which factors influence the effectiveness mostly. This should enhance the efficiency of advertising agents and optimize the advertisers’ advertising cost.

2. Summary of literature

The early papers mainly focus on targeted advertising on traditional media such as television (Gal-Or et al., 2006) and newspapers (Chandra, 2009). With the decline of traditional media, researchers now pay more attention to online targeted advertising. Many papers study the Online targeted advertising by experiments or surveys (Zhang et al., 2009; Bleier and Eisenbeiss, 2015; Barajas et al., 2016; Sahni et al., 2016; Liu and Mattila, 2017). Some other papers come up economic models to analyze the online advertising decision (e.g. Shin, 2015) and customers’ behavior towards online advertisements (e.g. Shen and Miguel Villas-Boas, 2017).

With the online advertising data, researchers can analyze the targeted advertising empirically. Multiple online advertising creatives should be major factors that impact the effects of online targeted advertising. Goldfarb and Tucker (2011) examine the effectiveness of different online display advertising strategies using data from a large randomized field experiment. They focus on two aspects of online advertising: visibility and contextual targeting, and use the variable whether the survey taker reported very likely to make a purchase when visit the website as the main dependent variable. Braun and Moe (2013) are also interested in the effect of different creatives, but they just pay attention to an advertising campaign run by a single automobile brand rather than multi product brands in Braun and Moe’s research. The paper also has more detailed classification of the advertising effectiveness (impressions, visits, and conversions).

What’s more, Andrews et al. (2016) analyze customers’ online purchase behavior when they are faced with physical crowdedness using mobile ads data provided by a China mobile operator. This study shows that customers’ offline environment also matters when making an online advertising decision, which is different from those papers concentrate on online factors only.

The measurement of the advertising effectiveness is also an important aspect. Wu (2015) compares the different advertising networks between Taobao/Alimama (decentralized) and Google (centralized). A structural model is constructed to maximize the advertiser’s profit, which, instead of visitors’ direct reaction towards advertisements, is measured as the effect of the online advertising. Advertising agents can match the advertisers’ demand with the publishers’ supply. However, few papers like Wu (2015) give attention to the role of advertising agents played in targeted advertising on social media.

3. Preliminary ideas

The data that we may use can be divided into four levels, and data in level 1 to level 3 are dependent variables and data in level 4 can be used as independent variables.

(1) Products/Goods: brand, classification, sale price, advertising budget (bid price for advertisement)

Brand, classification, and sale price are very important control variables to avoid endogeneity. If the advertisers’ bid price for advertisement and the publishers’ offer price are both available, a research the matching between these two prices will be very interesting. Also, we can figure out how the advertisers will behave with limited advertisement.

(2) Advertisement article: article framework (length, contents), creatives (pictures, videos, H5 pages, mini games), post frequency and timing

Nowadays, advertisers are trying to develop more and more fancy styles to promote their products. However, will those potential customers be attracted by this? It should be figured out that the simpler the better or the fancier the better.

(3) Platform/Official account: classification, number of fans/active fans, account age, followers’ tags (age, gender, location), offer price

Considering the matching of characteristics of the products and the publishers should be a central task of targeted advertising on WeChat platform. The followers of an account not always share the same tags or interests.

(4) Effectiveness: product sales, advertisement article (number of reading/Like, number of Like of the first comment)

Product sales may be difficult to obtain, but data of the advertisement articles after posting should be much easier to obtain. The number of reading measures how many readers click to open the article, but it does not mean they finish this article. Because the button “Zan” (“Like” in Chinese) is at the bottom of the article; the number of “Zan” measures how many readers finish the article and like this article; the number of “Zan” of the first comment reflects that how many readers are impressed by the first comment written by other reader. Therefore, we can observe the effectiveness of the online targeted advertising at three levels. This can help us avoid the misleading of the publishers’ exaggeration of the account’s influence.

4. Reference

Andrews, M., Luo, X., Fang, Z., & Ghose, A. (2015). Mobile ad effectiveness: Hyper-contextual targeting with crowdedness. Marketing Science, 35(2), 218-233.

Barajas, J., Akella, R., Holtan, M., & Flores, A. (2016). Experimental designs and estimation for online display advertising attribution in marketplaces. Marketing Science, 35(3), 465-483.

Bleier, A., & Eisenbeiss, M. (2015). Personalized online advertising effectiveness: The interplay of what, when, and where. Marketing Science, 34(5), 669-688.

Braun, M., & Moe, W. W. (2013). Online display advertising: Modeling the effects of multiple creatives and individual impression histories. Marketing Science, 32(5), 753-767.

Bruce, N. I., Murthi, B. P. S., & Rao, R. C. (2017, March). A Dynamic Model for Digital Advertising: The Effects of Creative Format, Message Content, and Targeting on Engagement. American Marketing Association.

Chandra, A. (2009). Targeted advertising: The role of subscriber characteristics in media markets. The Journal of Industrial Economics, 57(1), 58-84.

Gal-Or, E., Gal-Or, M., May, J. H., & Spangler, W. E. (2006). Targeted advertising strategies on television. Management Science, 52(5), 713-725.

Goldfarb, A., & Tucker, C. (2011). Online display advertising: Targeting and obtrusiveness. Marketing Science, 30(3), 389-404.

Hartmann, W. R. (2010). Demand estimation with social interactions and the implications for targeted marketing. Marketing science, 29(4), 585-601.

Johnson, J. P. (2013). Targeted advertising and advertising avoidance. The RAND Journal of Economics, 44(1), 128-144.

Liu, S. Q., & Mattila, A. S. (2017). Airbnb: Online targeted advertising, sense of power, and consumer decisions. International Journal of Hospitality Management, 60, 33-41.

Sahni, N. S., Zou, D., & Chintagunta, P. K. (2016). Do targeted discount offers serve as advertising? Evidence from 70 field experiments. Management Science.

Schumann, J. H., von Wangenheim, F., & Groene, N. (2014, January). Targeted online advertising: Using reciprocity appeals to increase acceptance among users of free web services. American Marketing Association.

Shen, Q., & Miguel Villas-Boas, J. (2017). Behavior-Based Advertising. Management Science.

Shin, W. (2015). Keyword search advertising and limited budgets. Marketing Science, 34(6), 882-896.

Wu, C. (2015). Matching value and market design in online advertising networks: An empirical analysis. Marketing Science, 34(6), 906-921.

Wu, C. (2015). Matching value and market design in online advertising networks: An empirical analysis. Marketing Science, 34(6), 906-921.

Zhang, J., Wedel, M., & Pieters, R. (2009). Sales effects of attention to feature advertisements: a Bayesian mediation analysis. Journal of Marketing Research, 46(5), 669-681.