Preventive Medicine Reports 24 (2021) 101520

Contents lists available at ScienceDirect

Preventive Medicine Reports

journal homepage: www.elsevier.com/locate/pmedr

Use of persuasive strategies in food advertising on television and on social

media in Brazil
Jéssica Moreira da Silva a, Michele Bittencourt Rodrigues a, Juliana de Paula Matos a,
Lais Amaral Mais b, c, Ana Paula Bortoletto Martins b, Rafael Moreira Claro a, Paula
Martins Horta a, *
a
Departamento de Nutrição, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, 30130-100, Brazil
b
Instituto Brasileiro de Defesa do Consumidor, São Paulo, 05002-000, Brazil
c
Faculdade de Saúde Pública, Universidade de São Paulo, São Paulo, 01246-904, Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: We analyzed the use of persuasive advertising strategies by 18 food brands on TV and Facebook, Instagram, and
Ultra-processed foods YouTube in Brazil in April 2018. Advertising strategies were investigated from three groups: power of advertising
Food advertising strategies (n = 10) (e.g., use of licensed character, celebrities, awards, etc), use of the prize offering (n = 9) (e.g.,
Internet
pay 2 take 3 or more, gifts or collectable, limited edition, etc), and use of brand benefit claims (n = 8) (e.g.,
Social media
Television
messages that exalt sensory-based characteristics such as flavor, taste, aroma and recommend how to use/
consume the product, etc). Almost 90% of the brands were ultra-processed foods producers and they carried 52
ads on TV and 194 posts on social media platforms. A higher frequency of the strategy ‘cartoon/company owned
character’ was found on TV ads (19.2%; p < 0.0001) in comparison to social media platforms (0% on the three
platforms) while the presence of ‘famous sportsperson/team’ prevailed on YouTube (41.4%) in comparison to TV
(19.2%), Facebook (10.9%) and Instagram (9.1%), p < 0.0001. On YouTube ads, the claims ‘sensory-based
characteristics’ (86.2%), ‘suggested use’ (51.7%), and ‘emotive claims’ (31.0%) were more commonly seen in
comparison to the other media, while the claims about ‘new brand developments’ (23.1%), ‘price’ (9.6%) and
‘suggesting to children and the whole family to use the advertised product’ (21.1%) prevailed on TV. Ultra-
processed food brands are the main food companies that advertise on Brazilian TV and social media and the
message transmitted by these brands varies in each media according to the advertising strategies that are used.

1. Introduction Highly persuasive communication has the potential to attract cus­

tomers and develop brand awareness and loyalty (Close and Ham,
Food choices are determined by a combination of individual factors, 2016), which can be achieved by using persuasive strategies, i.e., a set of
such as psychological and cultural issues, and environmental factors, actions to convince consumers to buy a product based not exclusively on
such as the characteristics of food availability, accessibility, and food rational factors, but especially on emotional and sentimental ones
advertising (Swinburn et al., 2013; Swinburn et al., 1999; Glanz et al., (Santana et al., 2020). Some examples of persuasive advertising strate­
2005; Downs et al., 2020). gies are advertisement repetition, product demonstration, peer popu­
Advertising is one of the marketing dimensions that is capable of larity appeal, celebrity endorsement, and premiums (Rozendaal et al.,
increasing brand awareness and purchase intention. The effectiveness of 2011). Although all individuals are influenced by the persuasive power
advertising depends on two important factors: the reach of the com­ of marketing communication, children and adolescents are the most
mercial message and the characteristics of the message such as providing vulnerable groups because they lack cognitive abilities in differentiating
of new information, useful advice to the individuals, the capacity to an informative tone from commercial content (Carter et al., 2011; Cal­
create curiosity about the brand or product, and the ability to generate vert, 2020).
enjoyment (Bronner and Neijens, 2006). Studies on food advertising are centered on food television (TV) and

* Corresponding author at: Av. Alfredo Balena 190, 30130-100, Escola de Enfermagem, 3◦ andar, sala 312, Belo Horizonte, Minas Gerais, Brazil.
E-mail address: paulamhorta@gmail.com (P.M. Horta).

https://doi.org/10.1016/j.pmedr.2021.101520
Received 22 January 2021; Received in revised form 2 August 2021; Accepted 8 August 2021
Available online 21 August 2021
2211-3355/© 2021 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
J.M. Silva et al. Preventive Medicine Reports 24 (2021) 101520

show a high number of unhealthy food advertisements (ads) (Cairns food chains (Burger King and McDonald’s), six of ultra-processed bev­
et al., 2013; Kelly et al., 2019; Soares Guimarães et al., 2020; Leite et al., erages (Coca-Cola; Dolly; Fanta; Novo Frisco; Pepsi; and Guaraná
2020). With the advent of the internet, other media have also been used Antártica), two of cookies and sweets (Oreo and Snickers), four of meats
to promote food, among which social media platforms stand out (Kelly and sausages (Friboi; Perdigão; Sadia; and Seara), and four of other food
et al., 2013). These platforms are highly accessed by individuals and in categories (coffee powder: Café Pilão; chocolate powder: Nescau;
2019 Facebook, Instagram, and YouTube had 2.14 billion, 1 billion and, margarine: Qualy; and fermented milk drink: Yakult).
2 billion active accounts, respectively (Statista, 2019; YouTube, 2020). Then, the Facebook, Instagram, and YouTube pages of these brands
In comparison to the traditional and mass advertising approach, social were identified by searching the name of each brand on the platforms.
media marketing is characterized by modernism and refers to a more Currently, the number of active accounts on Facebook Brazil is around
permeable, open, and flexible social media marketing culture (Felix 150 million, while Instagram and YouTube have a total of 106.5 million
et al., 2017). and 105 million respectively (NapoleonCat, 2021; Forbes 2021). Brand
In Brazil, almost 90% of all food ads announced on the open-TV are communication in social media can be any piece of brand-related
from ultra-processed foods (Soares Guimarães et al., 2020) and 96% of communication distributed via social media that enables internet users
them included one or more persuasive advertising strategies (Santana to access, share, engage with, add to, and co-create (Voorveld, 2019). In
et al., 2020). On social media, in Brazil, only a single study has inves­ our study, we are limiting our approach to the content (posts and videos)
tigated food ads on 16 ultra-processed food brands pages on Facebook published by the selected food brands on their official pages on social
during 2015 and posts contained user conversations, promotions, and media platforms in April 2018. The definition for one month of data
celebrities’ endorsements (Horta et al., et al., 2018). collection followed the same procedure used for food television moni­
This study aims to add knowledge about food advertising strategies toring, enabling comparisons. Other studies have also monitored food
in multiple media in Brazil and analyze, for the first time, the use of advertising on social media during this period (Busse, 2018; Buchanan
these strategies by food brands on TV and three platforms of social et al., 2018; Jaichuen et al., 2019). When visiting the brand pages, the
media. In the country, 96% of the households contain at least one TV set number of subscribers of each brand page on the platforms was also
(IBGE, 2018) and 75% of the adult population have accounts on social registered.
media and uses it for a mean of 3.5 h per day (IBOPE, 2018). Brazilian On Facebook, only the Yakult brand page was not found and the
children and adolescents also access social media content, among 24.3 Dolly, Maturatta Friboi, Sadia, Snikers, and Seara brands did not publish
million adolescents (9 to 17 years old) interviewed for a survey, 20 posts during the data collection period. On Instagram, the companies
million (82%) reported using social media platforms (Tic kids, 2019). Yakult and Frisco were not found and the companies Snickers, Seara,
Our results will help better understand the exposure of the Brazilian Oreo, Friboi, Dolly, Sadia, and Perdigão did not publish in this period.
population to the persuasive content of food advertising in the country. Regarding YouTube, all the selected companies had channels on the
platform; however, the companies Yakult, Dolly, Friboi, Nescau, Frisco,
2. Methods Sadia, Seara, and Snikers did not post videos in the analyzed period.
These losses indicate that our sample was composed mainly of brands
This is an observational study that analyzed the persuasive adver­ that are more present on social media in Brazil.
tising strategies used by food brands on TV and three social media In all the four studied media, according to the INFORMAS protocol,
platforms: Facebook, Instagram, and YouTube in Brazil. 27 persuasive advertising strategies were investigated and classified into
Facebook is a social media platform that allows people to build three major groups: power of advertising strategies (n = 10), use of the
public profiles and establish explicit connections with others in their prize offering (n = 9), and use of brand benefit claims (n = 8) (Kelly,
social media, Instagram is another social media platform that provides 2017). Some examples of the power of the advertising strategies are the
users with video- and photo-sharing possibilities, and YouTube is a use of licensed characters, celebrities, awards, etc; the premium offers
video-sharing platform/content community (Voorveld, 2019). category includes pay 2 take 3 or more, gifts or collectable, limited
First, we will describe the sample selection, that was based on a edition, and others; the brand benefit claims strategies contain messages
previous study that monitored Brazilian TV advertising (Soares that exalt sensory-based characteristics (flavor, taste, aroma), recom­
Guimarães et al., 2020). In this study, the programming of the three mend how to use/consume the product, etc. More details of each
open TV channels with the largest national audience was recorded persuasive advertising strategy are presented in Box 1.
during eight non-consecutive days of April 2018 (four weekend days and
four weekdays) from 6 am to midnight, totaling 432 h of recording Box 1
(Soares Guimarães et al., 2020). The data collection procedure followed Persuasive advertising strategies investigated.
the protocol for monitoring TV advertising proposed by INFORMAS
(International Network for Food and Obesity/non-communicable dis­ Variables Description
eases Research, Monitoring and Action Support) (Kelly, 2017; Soares Power of advertising
Guimarães et al., 2020). According to this protocol, food advertising Cartoon/Company owned Representation of the brand character (e.g., M&Ms)
character
monitoring should occur in March, April, or May to ensure compara­
Licensed character Representation of a licensed character (e.g., Dora
bility over time and to avoid seasonality effects in the results over time the explorer)
(Kelly, 2017). INFORMAS is a global network of organizations and re­ Amateur sportsperson Representation amateur sportsperson (e.g., a person
searchers of public interest that aims to monitor, compare, and support playing a sport)
actions related to the public and private sector, creating healthy eating Celebrity (non-sports) Representation of celebrities, such as musicians,
actresses and actors, and others (e.g., Jamie Oliver)
environments, reducing the incidence of obesity and chronic non-
Movie tie-in Mention of movies exhibited in the country (e.g.,
communicable diseases (NCD) (Swinburn et al., 2013). Shrek)
A total of 7,991 TV ads were registered, of which 922 were from food Famous sportsperson/team Representation of sports celebrities or teams (e.g.,
and/or beverage products that were carried out exclusively by 42 food Neymar)
Non-sports/historical Mention of commemorative dates (e.g., Christmas,
companies/brands (Soares Guimarães et al., 2020). Thus, a single brand
events/festivals New Years’ Eve, Easter, etc)
can have more than one advertisement being broadcast on Brazilian TV ‘For kids’ Image of a child, ‘great for school lunches’, ‘for
and a single advertisement can be displayed more than once during the school lunchboxes’
programming. We selected the brands responsible for 80% of the TV ads, Awards
in a way to reach a great part of food brands advertising on Brazilian TV. (continued on next page)
By this criteria, sample selection resulted in 18 brands, being two of fast-

2
J.M. Silva et al. Preventive Medicine Reports 24 (2021) 101520

Box 1 (continued ) repertoire (88.9%; n = 16); the remained 11.1% (n = 2) brands sell
Variables Description processed and minimally processed foods predominantly. During the
Awards of the announced products (e.g., Best Food
data collection period, the 18 brands broadcast 52 ads on the pro­
Award 2014, award winning, number one best- gramming of the three TV channels. On social media, the evaluated
selling’) pages published 194 posts: 110 on Facebook, 55 on Instagram, and 29 on
Sports events Representation of sports events (e.g., Olympics YouTube. The pages with the highest number of subscribers on Facebook
games and the FIFA World Cup, etc)
were McDonald’s (77,208,625), Fanta (19,891,239), and Coca-Cola
Premium offers
Game and app downloads Brand indications about games and apps to be (18,269,817), while on Instagram, McDonald’s, Coca-Cola, and Burger
downloaded (e.g., Happy Meal App (App King led this ranking: 1,766,338; 1,021,168 and 731,735 subscribers,
McDonald’s with games and activities) respectively. On YouTube, the Fanta’s channel had 647,275 subscribers,
Contests Contest proposed by the brand that involves McDonald’s’ 249,199, and Sadia’s 121,061 (Table 1).
individual subscription (e.g., Coca-Cola Holiday
2020 Instant Win and Sweepstakes)
Among the food companies, McDonald’s stood out for the greater
Pay 2 take 3 or other Offer that enables buying more products and paying number of ads in the four media: the ads of this brand represented 38.8%
lesser (e.g., Buy one pizza and get one free) of the total. Fanta, Qualy, Yakult, Burger King, and Coca-Cola were the
20% extra or other Offer that enables buying a higher content of the brands that followed McDonald’s in the number of ads (Table 2).
product at the same price that the original package
In the four media, the advertising strategies ‘non-sports celebrities’,
(e.g., Kinder Bueno with 1 bar free)
Limited edition Offer of a specific edition of the product (e.g, ‘sports events’ and ‘famous sportsperson or team’ were the most used in
McDonald’s World Cup burger) the power of ads strategy category and this group of advertising strategy
Social charity Mention of social charity actions conducted by the prevailed on YouTube (89.7%) and on Instagram (72.7%) in comparison
brand (e.g., Ronald McDonald House Charities) to the other two media (TV: 65.3%; Facebook: 62.7%), p = 0.038. Also, a
Gift or collectable Offer of gifts or collectable associated with the
products (e.g., Kinder Joys)
higher frequency of the strategy ‘cartoon/company owned character’
Price discount Offer of the product with reduction price (e.g., 25% was found on TV ads (19.2%; p < 0.0001) in comparison to social media
off or low price) platforms (0% on the three platforms) while the presence of ‘famous
Loyalty program Offer of a loyalty program (e.g., rewards program, sportsperson/team’ prevailed on YouTube (41.4%) in comparison to TV
loyalty card)
(19.2%), Facebook (10.9%) and Instagram (9.1%), p < 0.0001 (Table 3).
Brand benefit claims
Sensory-based characteristics Exaltation of characteristics of the brand’s products In the premium offers’ category, in the four media, ‘limited editions’
such as taste, texture, appearance, aroma (e.g., and ‘gifts and collectibles’ were the most common strategies employed
M&M: melts in your mouth, not in your hand). on the ads. In addition, a higher frequency of ‘price discounts’ was noted
New brand development New product launched by the brand (e.g., new on TV (9.6%) and Facebook (9.1%) ads in comparison to Instagram
flavor)
Suggested use Indication of how to use the product (e.g., great for
(0.0%) and YouTube (0.0%) posts (p = 0.040) (Table 3).
lunchboxes) Finally, messages emphasizing sensory-based characteristics of the
Suggested users are children Indication to consume the products by children or advertised product (67.3%) and suggesting how to use the product
or whole family the whole family (29.3%) were the advertising strategies under the group of brand benefit
Emotive claims Message of fun, feelings, popularity (e.g., Open
claims that most prevailed in the ads in the four media. On YouTube ads,
happiness)
Puffery Claim to be advantageous over other products (e.g., the claims ‘sensory-based characteristics’ (86.2%), ‘suggested use’
the friendliest drink on the earth) (51.7%) and, ‘emotive claims’ (31.0%) were more commonly seen in
Convenience Messages about how easy is to acquire, prepare or comparison to the other media, while the claims about ‘new brand
consume the products of the brand (e.g., ready to
bake)
Price Claims about the beneficial aspects involving the Table 1
product cost Number of subscribers in each social media platform (Facebook, Instagram, and
YouTube) according to food brands. Brazil, April 2018.
Brand Facebook Instagram YouTube
Source: Kelly (2017).
Although the persuasive advertising strategies investigated were Fast food chains
Burger King 8,192,901 731,735 37,432
specifically proposed for monitoring TV advertising, it was used in the
McDonald’s 77,208,625 1,766,338 249,199
present study on social media platforms for purposes of standardization
and comparability of the studied media. Ultra-processed beverages
Coca-Cola 18,269,817 1,021,168 10,725
Data were tabulated using the Microsoft® Excel 2010 program and Dolly 160,278 * *
the public domain software Epi-info (StataCorp LP, College Station, Fanta 19,891,239 * 647,275
United States). All data were tabulated independently by two re­ Novo Frisco 150,034 ** 351
searchers and cross-checked to correct any error (inter-coder reliability Pepsi 3,648,941 66,380 *
Guaraná Antártica 15,820,769 262,266 *
was ≥ 90%, except for three persuasive advertising strategies: sports
events (89.2%); sensory-based characteristics (72.2%), suggested use Cookies and sweets
(77.3%)). Despite this, all inconsistencies between both datasets were Oreo * * *
Snickers 10,446,358 * *
subsequently verified and resolved with a third researcher. Frequency
Nescau 2,260,411 12,068 32,633
estimates for the persuasive advertising strategies were calculated in the
whole sample and stratified by brand and type of media. Chi-squared Meat and sausages
Friboi 1,177,612 * 16,613
and Fisher’s Exact tests were applied to compare proportions of adver­ Perdigão 893,724 * 22,799
tising strategies among the four media using 5% as the significance level Sadia 3,794,262 64,582 121,061
(p-value < 0.05). The Stata statistical software version 12.1 was used for Seara 1,115,033 5,639 40,296
data analysis. Others
Café Pilão 275,196 5,000 13,012
3. Results Qualy 55,286 * *
Yakult ** ** *

The sample of brands included in the study was mainly represented *Number of subscribers not available.
by brands that contain a high proportion of ultra-processed foods in their **Brand has no page on the social media platform.

3
J.M. Silva et al. Preventive Medicine Reports 24 (2021) 101520

Table 2
Frequency of food advertising on television and social media (Facebook, Instagram, and YouTube) according to food brands. Brazil, April 2018.
Brand Total TV Facebook Instagram YouTube
(n = 246) (n = 52) (n = 110) (n = 55) (n = 29)

n % n % n % n % n %

Fast food chains

Burger King 19 7.8 3 5.8 11 10.0 2 3.6 3 10.3
McDonald’s 95 38.8 15 28.8 46 41.8 20 36.4 14 48.3

Ultra-processed beverages
Coca-Cola 17 6.9 2 3.8 8 7.3 6 10.9 1 3.5
Dolly 3 1.2 3 5.8 0 0.0 0 0.0 * *
Fanta 22 9.0 4 7.7 6 5.5 9 16.4 3 10.3
Novo Frisco 6 2.4 2 3.8 4 3.6 * * * 0.0
Pepsi 8 3.3 2 3.8 2 1.8 3 5.5 1 3.5
Guaraná Antártica 7 2.9 4 7.7 1 0.9 1 1.8 1 3.5

Cookies and sweets

Oreo 5 2.0 1 1.9 3 2.7 0 0.0 1 3.5
Snickers 1 0.4 1 1.9 0 0.0 0 0.0 * *
Nescau 13 5.3 1 1.9 9 8.2 3 5.5 * 0.0

Meat and sausages

Friboi 2 0.8 2 3.8 0 0.0 0 0.0 * *
Perdigão 8 3.3 3 5.8 3 2.7 0 0.0 2 6.9
Sadia 3 1.2 3 5.8 0 0.0 0 0.0 * *
Seara 1 0.4 1 1.9 0 0.0 0 0.0 * *

Others
Café Pilão 15 6.1 2 3.8 9 8.2 3 5.5 1 3.5
Qualy 20 8.2 2 3.8 8 7.3 8 14.6 2 6.9
Yakult 19 7.8 1 1.9 * * * * * *

Note: Data collection involved the three open TV channels with the largest national audience. Programming of eight non-consecutive days of April 2018 were recorded
from 6:00 am to midnight. On the internet, data collection included all social media posts published by the brands during April 2018. *Brand has no page on the social
media platform.

Table 3
Frequency of persuasive advertising strategies used by the food brands on television and on social media (Facebook, Instagram, and YouTube). Brazil, April 2018.
Persuasive marketing strategy Total - % Media – % p-value
(n = 246)
TV Facebook Instagram YouTube
(n = 52) (n = 110) (n = 55) (n = 29)

Power of advertising 100.0 65.3 62.7 72.7 89.7 0.038

Celebrity (non-sports) 38.0 13.5 19.1 30.9 31.0 0.083
Cartoon/Company owned character 4.1 19.2 0.0 0.0 0.0 <0.0001
Sports events 42.8 21.1 26.4 29.1 41.4 0.264
‘For kids’ 14.0 19.2 9.1 7.3 3.5 0.080
Licensed character 12.8 11.5 9.1 7.3 6.9 0.856
Famous sportsperson/team 20.6 19.2 10.9 9.1 41.4 <0.0001
Amateur sportsperson 2.4 3.8 1.8 0.0 6.9 0.221
Non-sports/historical events/festivals 0.8 1.9 0.9 0.0 0.0 0.683
Movie tie-in 0.0 0.0 0.0 0.0 0.0 *
Awards 0.0 0.0 0.0 0.0 0.0 *
Premium offers 71.9 42.3 53.6 43.6 55.2 0.399
Price discount 6.1 9.6 9.1 0.0 0.0 0.040
Contests 7.5 5.8 10.0 1.8 3.5 0.196
Limited edition 41.6 26.9 21.8 29.1 37.9 0.336
Gift or collectable 19.6 15.4 12.7 12.7 10.3 0.929
Loyalty programs 5.2 0.0 4.6 3.6 3.5 0.501
Pay 2 take 3 or other 1.2 0.0 2.7 0.0 0.0 0.289
Game and app downloads 0.4 0.0 0.9 0.0 0.0 0.743
20% extra or other 5.6 0.0 4.6 3.6 6.9 0.379
Social charity 0.0 0.0 0.0 0.0 0.0 *
Brand benefit claims 89.7 82.7 60.9 45.5 93.1 <0.0001
Sensory-based characteristics 67.3 59.6 50.9 29.1 86.2 <0.0001
Suggested use 29.7 9.6 20.9 16.4 51.7 <0.0001
New brand development 9.8 23.1 7.3 0.0 13.8 <0.0001
Price 6.1 9.6 9.1 0.0 0.0 0.040
Puffery 0.4 0.0 0.9 0.0 0.0 0.743
Suggested users are children or family 14.4 21.1 8.2 7.3 6.9 0.049
Convenience 2.1 0.0 4.6 0.0 0.0 0.097
Emotive claims 15.0 0.0 7.3 10.9 31.0 <0.0001

Note: Data collection involved the three open TV channels with the largest national audience. Programming of eight non-consecutive days of April 2018 were recorded
from 6:00 am to midnight. On the internet, data collection included all social media posts published by the brands during April 2018.
*Frequency of the marketing strategy was equal to zero in all media.

4
J.M. Silva et al. Preventive Medicine Reports 24 (2021) 101520

developments’ (23.1%), ‘price’ (9.6%) and ‘suggesting to children and as the most common marketing strategy, followed by the premium offers
the whole family to use the advertised product’ (21.1%) prevailed on (Vandevijvere et al., 2018). On Facebook in Thailand, all posts made by
TV. food brands contained at least one persuasive marketing strategy being
the most frequent: the use of images, brand elements, and hashtags
4. Discussion (Jaichuen et al., 2019). On Australian Instagram pages, 15 food brands
showed athletes frequently on posts (Vassallo et al., 2018). And in New
Our results showed particularities in the use of persuasive adver­ Zealand, 61% of 300 videos published by 15 food brands contained
tising strategies on TV and in three social media platforms in Brazil. discounts, and 24% offered gifts (Vandevijvere et al., 2018). In Malaysia,
Although the high occurrence of these strategies was expected on the videos of YouTube aimed at children contained claims of appealing
ads, since advertising content is always persuasive, this study high­ benefits to taste (42%), new product promotion (32%), use of anima­
lighted the nuances about the commercial message that was being tions (22%), promotions (15%), and health and nutrition (8%) (Tan
transmitted in each media. Another important finding of the study is that et al., 2018).
almost 90% of the brands that most carried out ads on Brazilian TV in Our study innovates by revealing the occurrence of advertising
2018 were from the ultra-processed food segment. The majority of these strategies in different media which allowed us to better understand how
brands have pages on social media and use the platforms to announce food advertising is aligned with the characteristics of each media.
their products and interact with individuals. Although digital technology has developed in the last decades, TV is still
The high presence of ultra-processed food advertising on TV has the main consumed media in many parts of the world and Brazil (EGTA,
already been pointed out by other studies in different countries (Kelly 2018). Also, an innovative study compared the effects of individuals
et al., 2019; Kelly et al., 2010). In 2010, a collaboration between 13 being exposed to ads either on TV programs or videos on YouTube. The
research groups showed a high volume of TV advertising for unhealthy authors found that TV leads to more attention and to more positive
foods in the global programming aimed at children (Kelly et al., 2010). emotions advertising than online advertising which means that TV
More recently, data from INFORMAS network described the prevalence advertising has a stronger immediate impact on the recipient (Weibel
of food advertising on TV in 22 countries, corresponding to 23% of the et al., 2019). The authors concluded that TV advertising is still the better
total number of ads recorded in 11,191 h of broadcasting (Kelly et al., choice for advertisers (Weibel et al., 2019). This finding can explain the
2019). high frequency of persuasive advertising found on TV ads that guaran­
In Brazil, a study showed the occurrence of 10.8% of food ads on tees a commercial message that attracts consumers.
Brazilian open TV in 2018, 90.8% of which were ultra-processed foods On social media, brands use marketing resources to stimulate sales,
(Soares Guimarães et al., 2020). In addition, results have shown that increase brand awareness, improve brand image, generate traffic to
80% of the foods advertised did not meet the nutritional profile criteria online platforms, reduce marketing costs, and create user interactivity
of the Pan American Health Organization (PAHO) and the World Health on platforms by stimulating users to post or share content (Felix et al.,
Organization (WHO), recommended for use in restrictive advertising 2017). In these media, marketers can deliver different messages tailored
measures worldwide (Leite et al., 2020). to individual consumers. Based on what marketers know about them,
In social media, other investigations have also demonstrated a large some individuals will be offered rewards, discounts, or information;
investment by ultra-processed food brands in food advertising in others might be viewed as having a low lifetime-revenue potential and
Thailand, New Zealand, and Australia (Jaichuen et al., 2019; Vande­ given less favorable treatment or ignored entirely (Montgomery et al.,
vijvere et al., 2018; Vassallo et al., 2018). In Brazil, 16 pages of ultra- 2019).
processed food brands were among the 250 most liked pages by Bra­ However, social media should not be treated as an umbrella concept,
zilians on Facebook in 2014–2015 (Horta et al., 2018). No other study since not all platforms are created equal and different functionalities and
has monitored food brands pages on Instagram and YouTube in Brazil. characteristics of social media platforms translate into different con­
Besides the widespread of ultra-processed food advertisements in sumer experiences (Voorveld, 2019). For example, Facebook and
multiple media in Brazil, it is also important to address the potential Instagram are mainly used for social interaction, to fill an empty
reach of these ads. Almost all households (96.4%) in Brazil have at least moment, and to be quickly informed and up to date. On the other hand,
one TV set (IBGE, 2018). This led to great number of individuals users choose YouTube for entertainment purposes and to feel happy,
watching the TV channels considered during the period of data collec­ relaxed, and to have a moment for themselves (Voorveld et al., 2018).
tion. These individuals have thus been exposed to the commercial The different social media proposals and the users’ perceptions about
message present in the ads. Regarding social media, the ads posted by them can help explain the presence of different persuasive advertising
the brands on their official pages have the potential to reach at least all strategies in the food ads. Ads on YouTube channels presented more
the subscribers of the pages, which represents more than 50,000,000 strategies that enhanced the ad power and strategies that highlighted
individuals. The advertising reach in social media can be even higher positive characteristics of the product and the brand, e.g., brand benefit
and ads can also be seen by other social media users. This happens claims. To be effective, advertisements on this platform should be pos­
because companies have been using user data available on the internet itive, fun, and interesting, otherwise, individuals can find them intrusive
to target advertisements based on the type of content consumed by the and annoying, making them wait several seconds to be able to skip the
users (Montgomery et al., 2019). Since these messages are tailored to ad and continue watching the video content (Voorveld et al., 2018).
individual consumers, the potential persuasion of the exposure of this On Facebook and Instagram, advertisements can be also disturbing
type of advertising is higher than the non-tailored ones. since an ad placed in someone’s timeline interrupts what is meant to be a
Apart from reaching a lot of individuals, the ads we have studied also social experience, which is the main purpose of these platforms. This
contained different advertising strategies to attract consumers. Previous causes consumers to view the ad in a negative light, with possible
studies have also addressed this aspect. On TV, a systematic review of 38 drawbacks for the advertiser. Although the evaluated food brands have
articles showed a predominance of the premium offers, promotional posted more times on Facebook and Instagram during the data collection
characters, nutritional claims, and appeal for taste and fun in this media period, the investment in using the adverting strategies seems to have
(Jenkin et al., 2014). The recent study of monitoring of TV advertising in been less important.
22 countries also highlighted the use of promotional characters and Another aspect that is important to be discussed in the social media
offering prizes as frequent advertising strategies on TV (Kelly et al., marketing theme is the audience of the platforms and access to their
2019). content. In Brazil, and other parts of the world, social media platforms
On social media, the monitoring of food advertising on Facebook in are mainly accessed by young adults (18–44 years old) (Statista,2019)
New Zealand revealed the endorsement by a famous sportsman or team and social media platforms restrict registration to individuals under 13

5
J.M. Silva et al. Preventive Medicine Reports 24 (2021) 101520

years old. Although individuals of all ages are influenced by the unhealthy foods ads is essential for the country’s food information
persuasive content of advertising (Vukmirovic, 2015), children are the environment to promote healthy food choices, without the influence of
most vulnerable subjects (Carter et al., 2011; Calvert, 2020). However, unhealthy food companies.
we can not affirm that only individuals above 13 years old are exposed to Although being pioneering research that describes and compares
food advertising on social media. When registering on these platforms, food advertising carried out in Brazil on TV and social media platforms,
children can misinform their real age and get inside the social media this study has limitations that need to be addressed. First, we selected
universe. In the case of YouTube, in order to protect children from the marketing strategies to be studied based on INFORMAS (Kelly, 2017)
abusive content, the company created a specific platform for their ac­ protocol, an instrument specifically proposed to monitor food adver­
cess, known as YouTube Kids. But the content of all channels of YouTube tising on TV. This instrument contains the main persuasive marketing
can be accessed by any individual without log-in. strategies used by the food industry to advertise their products and in­
According to the above, we can conclude that Brazilian children are fluence food consumption. Although we have not measured the impact
potentially exposed to the food advertising content and our results of the individual exposure to advertisements containing these strategies,
showed a high prevalence of advertising strategies that dialogues literature has shown the high impact of such exposure in persuading
directly with the children (e.g., cartoon/company owned character; consumers of all ages. Second, despite the risk of not being able to
licensed character; ‘For kids’; suggested users are children or whole capture particularities of social media, due to the INFORMAS (Kelly,
family, among others), especially on TV and YouTube. This makes 2017) protocol being an instrument developed to monitor TV ads, the
important the discussion about the abusiveness of the advertising in same was applied in the present study of digital media. This choice was
Brazil. made due to the objective of comparing the findings between TV and the
The Consumer Defense Code (Código de Defesa do Consumidor - CDC) internet. Third, we selected the most popular social media platforms in
of 1990 considers illegal all the ads that transfer false information or any Brazil: the percentage of Brazilians that use Facebook, Instagram, and
content capable of inducing the subject to error, in addition to the ads Youtube are above 40%. However, other platforms also contain food
that use the deficiency of judgment and inexperience of a child or advertising but are lesser important in the context of Brazil. For
propagate information capable of inducing the consumer to a behavior example, Twitter and Tumblr are used by less than 15% of the popula­
that can damage his health or safety (Henriques et al., 2014). Also, the tion. In addition, some brands that composed our TV sample did not
Resolution 163/2014 of the National Council for the Rights of Children have accounts on social media or have not been published during the
and Adolescents (Conselho Nacional dos Direitos da Criança e do Ado­ data collection period. Also, the study was carried out in 2018, when
lescente – CONANDA) describes the types of abusive advertising directed occurred the FIFA World Cup. Consequently, marketing strategies
at children that intend to persuade them to ingest or buy any products or related to sports events were the most common in the sample, which can
services, whatever the means of communication that is transmitted not represent the food advertising profile outside these dates.
(Henriques et al., 2014).
In addition to state regulation, the National Council for Advertising 5. Conclusion
Self-Regulation (Conselho Nacional de Autorregulamentação Publicitária –
CONAR), formed by entities linked to advertising production and in­ The study showed, for the first time, characteristics of persuasion of
dustry, proposes restrictions on food advertising and advertising aimed food advertising in Brazil by applying a comparison between TV and
at children (CONAR, 2021). However, this model of regulation is criti­ three social medial platforms. We found that most brands that announce
cized from the point of view of representativeness of the interests on Brazilian TV are ultra-processed food producers and have pages on
involved since there is no representation of civil society and the State in social media where they use a varied repertoire of food advertising
the composition of the council (Henriques et al., 2014). strategies. The implications of our findings refer to the importance of
Despite the content of the CDC, the CONANDA Resolution, and the regularly monitoring food advertising on different media and studying
self-regulation proposed by CONAR, the weakness of these regulations is the impact of individual exposure to the persuasive content, specially on
exposed by the results of the present study. First, Brazil’s federal legis­ the digital landscape where literature is lack, in a way to guide consumer
lation must be appropriately overseen and penalties should be enforced defense actions and revisions of legislation.
in case of breaches (Henriques et al., 2014). In the self-regulating field,
penalties are not sufficiently imposed (CONAR, 2021). In the digital CRediT authorship contribution statement
landscape, low levels of monitoring can be explained by the necessity of
high-tech tools to collect data about food advertising on the internet. Jéssica Moreira Da Silva: Conceptualization, Data curation, Formal
Also, social media allow ordinary people, who are not part of the food analysis, Writing - original draft. Michele Bittencourt Rodrigues: Data
industry, to replicate advertisements by sharing or retweeting them on curation, Formal analysis, Writing - original draft. Juliana de Paula
their social media. The easy and speedy sharing of these advertisements Matos: Conceptualization, Data curation, Formal analysis, Writing -
makes it difficult to trace and attribute responsibility. The process of review & editing. Laís Amaral Mais: Conceptualization, Methodology,
identifying the advertisements is also hampered by sponsorship strate­ Writing - review & editing. Ana Paula Bortoletto Martins: Conceptu­
gies involving digital influencers or celebrities. Companies have alization, Funding acquisition, Methodology, Writing - review & editing.
consistently sponsored these influencers to advertise their products on Rafael Moreira Claro: Conceptualization, Methodology, Writing - re­
social media, frequently hiding the commercial partnership behind it view & editing. Paula Martins Horta: Conceptualization, Formal
(Rodrigues et al., 2021; UNSCN, 2020). analysis, Methodology, Writing - review & editing.
Together, these aspects reinforce the importance of strengthening
advertising legislation in Brazil, including the applicability to the digital Declaration of Competing Interest
landscape. This can be achieved by creating channels for registering
complaints, and also by speeding up the process of dealing with these The authors declare that they have no known competing financial
complaints. The food industry must comply with the regulatory terms in interests or personal relationships that could have appeared to influence
their ads and the public authorities must endeavor to inspect their ac­ the work reported in this paper.
tions. Scientists should invest in research on the subject with the aim of
systematically monitoring advertising practices in the food industry and Acknowledgment
their impact on consumer health. Civil society can be motivated through
campaigns to raise public awareness regarding the issue of food adver­ This work was supported by the IDRC [grant number 108166-001]
tising (Rodrigues et al., 2021). Reducing the population’s exposure to and Bloomberg Philanthropies through a sub award agreement [grant

6
J.M. Silva et al. Preventive Medicine Reports 24 (2021) 101520

number: 5104695] between the University of North Carolina at Chapel Cauchi, D., Correa, T., Corvalán, C., Cosenza-Quintana, E.L., Fernández-Escobar, C.,
González-Zapata, L.I., Halford, J., Jaichuen, N., Jensen, M.L., Karupaiah, T.,
Hill and the Fundação Faculdade de Medicina, Brazil.
Kaur, A., Kroker-Lobos, M.F., Mchiza, Z., Miklavec, K., Parker, W.-a., Potvin
Kent, M., Pravst, I., Ramírez-Zea, M., Reiff, S., Reyes, M., Royo-Bordonada, M.Á.,
References Rueangsom, P., Scarborough, P., Tiscornia, M.V., Tolentino-Mayo, L., Wate, J.,
White, M., Zamora-Corrales, I., Zeng, L., Swinburn, B., 2019. Global benchmarking
Bronner, F., Neijens, P., 2006. Audience experiences of media context and embedded of children’s exposure to television advertising of unhealthy foods and beverages
advertising: a comparison of eight media. Int. J. Market Res. 48 (1), 81–100. https:// across 22 countries. Obes. Rev. 20 (S2), 116–128. https://doi.org/10.1111/obr.v20.
doi.org/10.1177/147078530604800106. S210.1111/obr.12840.
Buchanan, L., Yeatman, H., Kelly, B., Kariippanon, K., 2018. A thematic content analysis Kelly, B., Halford, J.C.G., Boyland, E.J., Chapman, K., Bautista-Castaño, I., Berg, C.,
of how marketers promote energy drinks on digital platforms to young Australians. Caroli, M., Cook, B., Coutinho, J.G., Effertz, T., Grammatikaki, E., Keller, K.,
Aust. N. Z. J. Public Health 42 (6), 530–531. https://doi.org/10.1111/1753- Leung, R., Manios, Y., Monteiro, R., Pedley, C., Prell, H., Raine, K., Recine, E., Serra-
6405.12840. Majem, L., Singh, S., Summerbell, C., 2010. Television food advertising to children: a
Busse, P., 2018. Analysis of advertising in the multimedia environment of children and global perspective. Am. J. Public Health 100 (9), 1730–1736. https://doi.org/
adolescents in Peru. J. Children Media 12 (4), 432–447. https://doi.org/10.1080/ 10.2105/AJPH.2009.179267.
17482798.2018.1431557. Leite, F.H.M., Mais, L.A., Ricardo, C.Z., Andrade, G.C., Guimarães, J.S., Claro, R.M.,
Cairns, G., Angus, K., Hastings, G., et al., 2013. Systematic reviews of the evidence on the Duran, A.C.d.F.L., Martins, A.P.B., 2020. Nutritional quality of foods and non-
nature, extent and effects of food marketing to children. A retrospective summary. alcoholic beverages advertised on Brazilian free-to-air television: a cross-sectional
Appetite 62, 209–215. https://doi.org/10.1016/j.appet.2012.04.017. study. BMC Public Health 20 (1). https://doi.org/10.1186/s12889-020-08527-6.
Calvert, S.L., 2020. Children as consumers: advertising and marketing. Future Children Montgomery, K., Chester, J., Nixon, L., et al., 2019. Big Data and the transformation of
18 (1), 205–234. https://doi.org/10.1353/foc.0.0001. food and beverage marketing: undermining efforts to reduce obesity? Critical Public
Carter, O.B., Patterson, L.J., Donovan, R.J., Ewing, M.T., Roberts, C.M., 2011. Children’s Health. 29 (1), 110–117. https://doi.org/10.1080/09581596.2017.1392483.
understanding of the selling versus persuasive intent of junk food advertising: NapoleonCat, 2021. Social media users in Brazil. [Internet] Available in: https://na
implications for regulation. Soc. Sci. Med. 72 (6), 962–968. https://doi.org/ poleoncat.com/stats/social-media-users-in-brazil/2021/05 [Accessed: 14/06/
10.1016/j.socscimed.2011.01.018. 2021].
CONAR, 2021. Código Brasileiro de Autorregulamentação Publicitária [Brazilian Rodrigues, M., Matos, J., Horta, P., 2021. The COVID-19 pandemic and its implications
Advertising Self-Regulation]. Available in: http://www.conar.org.br/codigo/codigo. for the food information environment in Brazil. Public Health Nutr. 24 (2), 321–326.
php [Accessed: 14/06/2021]. https://doi.org/10.1017/S1368980020004747.
Close, J.P., Ham, J., 2016. Persuasive Communication. In: Close JP, editor. AiREAS: Rozendaal, E., Buijzen, M., Valkenburg, P., 2011. Children’s understanding of
Sustainocracy for a Healthy City: Phase 3: Civilian Participation – Including the advertisers’ persuasive tactics. Int. J. Advertis. 30 (2), 329–350. https://doi.org/
Global Health Deal Proposition [Internet]. Cham (CH): Springer. Chapter 1. doi: 10.2501/IJA-30-2-329-350.
10.1007/978-3-319-45620-1_1. Available in: https://www.ncbi.nlm.nih. Santana, M.O., Guimarães, J.S., Leite, F.H.M., et al., 2020. Analysing persuasive
gov/books/NBK500233/. marketing of ultra-processed foods on Brazilian television. Int. J. Public Health 65,
Downs, S.M., Ahmed, S., Fanzo, J., Herforth, A., 2020. Food environment typology: 1067–1077. https://doi.org/10.1007/s00038-020-01456-6.
advancing an expanded definition, framework, and methodological approach for Statista, 2019. Global social networks ranked by number of users. Available in: https
improved characterization of wild, cultivated, and built food environments toward ://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-o
sustainable diets. Foods 9 (4), 532. https://doi.org/10.3390/foods9040532. f-users/. [Accessed: 05/02/2020].
EGTA, European Group of Television Advertising, 2018. Advances in Hybrid Television Swinburn, B.A., Egger, G., Raza, F., 1999. Dissecting obesogenic environments: the
Audience Measurement [Internet]. Available in: http://www.egta.com/document development and application of a framework for identifying and prioritizing
s/2018_egta_insight_tam_EDITION_3 [Accessed in: 14/06/2021]. environmental interventions for obesity. Prev. Med. 29 (6), 563–570. https://doi.
Felix, R., Rauschnabel, P.A., Hinsch, C., 2017. Elements of strategic social media org/10.1006/pmed.1999.0585.
marketing: a holistic framework. J. Business Res. 70, 118–126. Swinburn, B., Sacks, G., Vandevijvere, S., Kumanyika, S., Lobstein, T., Neal, B.,
Forbes, 2021. YouTube lança ferramenta de vídeos curtos no Brasil. [YouTube launches Barquera, S., Friel, S., Hawkes, C., Kelly, B., L’Abbé, M., Lee, A., Ma, J.,
short videos tool in Brazil]. June. Available in: https://forbes.com.br/forbes-tech Macmullan, J., Mohan, S., Monteiro, C., Rayner, M., Sanders, D., Snowdon, W.,
/2021/06/youtube-lanca-ferramenta-de-videos-curtos-no-brasil/ [Accessed in: 14/ Walker, C., 2013. INFORMAS (International Network for Food and Obesity/non-
06/2021]. communicable diseases Research, Monitoring and Action Support): overview and
Jenkin, G., Madhvani, N., Signal, L., Bowers, S., 2014. Persuasive TV food marketing to key principles. Obes. Rev. 14, 1–12. https://doi.org/10.1111/obr.12087.
children. Obes. Rev. 15, 281–293. https://doi.org/10.1111/obr.12141. Tan, L., Ng, S.H., Omar, A., Karupaiah, T., 2018. What’s on YouTube? A case study on
Glanz, K., Sallis, J.F., Saelens, B.E., Frank, L.D., 2005. Healthy nutrition environments: food and beverage advertising in videos targeted at children on social media. Child
concepts and measures. Am. J. Health Promotion 19 (5), 330–333. https://doi.org/ Obes. 14 (5), 280–290. https://doi.org/10.1089/chi.2018.0037.
10.4278/0890-1171-19.5.330. TIC Kids, 2019. Pesquisa sobre o uso da Internet por crianças e adolescentes no Brasil:
Soares Guimarães, J., Mais, L., Marrocos Leite, F., Horta, P., Oliveira Santana, M., TIC Kids Online Brasil [livro eletrônico] = Survey on Internet use by children in
Martins, A., Claro, R., 2020. Ultra-processed food and beverage advertising on Brazil: ICT Kids Online Brazil 2019 / [editor] Núcleo de Informação e Coordenação
Brazilian television by International Network for Food and Obesity/Non- do Ponto BR. – 1. ed. – São Paulo: Comitê Gestor da Internet no Brasil, 2020.
Communicable Diseases Research, Monitoring and Action Support benchmark. Available in: https://nic.br/media/docs/publicacoes/2/20201123093344/tic_kids
Public Health Nutr. 23 (15), 2657–2662. https://doi.org/10.1017/ _online_2019_livro_eletronico.pdf [Accessed: 12/11/2020].
S1368980020000518. UNSCN N, 2020. Nutrition 45: Nutrition in a Digital World [Internet]. Jul [cited 2021
Henriques, P., Dias, P.C., Burlandy, L., 2014. Regulation of food advertising in Brazil: June 14]. Available in: https://www.unscn.org/uploads/web/news/UNSCN
convergence and conflicts of interest. Cad Saúde Públ. 30 (6), 1219–1228. https:// -Nutrition-45-WEB.pdf.
doi.org/10.1590/0102-311X00183912. Vandevijvere, S., Aitken, C., Swinburn, B., 2018. Volume, nature and potential impact of
Horta, P., Rodrigues, F., Dos Santos, L., 2018. Ultra-processed food product brands on advertisements on Facebook and YouTube by food brands popular in New Zealand.
Facebook pages: highly accessed by Brazilians through their marketing techniques. N Z Med J. April, 131 (1473): 14-24. Available in: https://www.nzma.org.nz/journa
Public Health Nutr. 21 (8), 1515–1519. https://doi.org/10.1017/ l-articles/volume-nature-and-potential-impact-of-advertisements-on-facebook-and-
S1368980018000083. youtube-by-food-brands-popular-in-new-zealand.
IBGE, 2018. Diretoria de Pesquisas, Coordenação de Trabalho e Rendimento. Pesquisa Vassallo, Amy Jo, Kelly, Bridget, Zhang, Lelin, Wang, Zhiyong, Young, Sarah,
Nacional por Amostra de Domicílios Contínua 2017-2018 [Continuous National Freeman, Becky, 2018. Junk food marketing on instagram: content analysis. JMIR
Household Sample Survey 2017-2018]. Available in: https://biblioteca.ibge.gov. Public Health Surveill. 4 (2), e54. https://doi.org/10.2196/publichealth.9594.
br/visualizacao/livros/liv101705_informativo.pdf [Accessed: 14/06/2020]. Voorveld, H.A.M., Noort, G.V., Muntinga, D.G., Bronner, F., 2018. Engagement with
IBOPE, 2018. Apps de redes sociais são os mais usados em smartphones. [Social media social media and social media advertising: the differentiating role of platform type.
apps are the most used on smartphones]. Available in: http://ibopeconecta.com/app J. Advertis. 47 (1), 38–54. https://doi.org/10.1080/00913367.2017.1405754.
s-de-redes-sociais-sao-os-mais-usados-em-smartphones/. [Accessed: 22/03/2020]. Voorveld, H.A.M., 2019. Brand communication in social media: a research agenda.
Jaichuen, N., Vongmongkol, V., Suphanchaimat, R., Sasiwatpaisit, N., J. Advertis. 48 (1), 14–26. https://doi.org/10.1080/00913367.2019.1588808.
Tangcharoensathien, V., 2019. Food marketing in facebook to thai children and Vukmirovic, M., 2015. The effects of food advertising on food-related behaviours and
youth: an assessment of the efficacy of thai regulations. Int. J. Environ. Res. Public perceptions in adults: a review. Food Res. Int. 75, 13–19. https://doi.org/10.1016/j.
Health 16 (7), 1204. https://doi.org/10.3390/ijerph16071204. foodres.2015.05.011.
Kelly, B., King, L., Baur, L., et al., 2013. Monitoring food promotions to children. Obes. Weibel, D., di Francesco, R., Kopf, R., et al., 2019. TV vs. YouTube: TV advertisements
Rev. 14, 59–69. https://doi.org/10.1111/obr.12076. capture more visual attention, create more positive emotions and have a stronger
Kelly, B., 2017. Food Marketing – Television: INFORMAS Food Promotion Protocol impact on implicit long-term memory. Front. Psychol. 10, 626. https://doi.org/
[Internet]. University of Auckland. Available in: informas.org [Accessed: 12/11/ 10.3389/fpsyg.2019.00626.
2020]. YouTube, 2020. YouTube para a imprensa [YouTube for the press]. Available in: https://
Kelly, B., Vandevijvere, S., Ng, SeeHoe, Adams, J., Allemandi, L., Bahena-Espina, L., www.youtube.com/intl/pt-BR/about/press/. [Accessed: 03/04/2020].
Barquera, S., Boyland, E., Calleja, P., Carmona-Garcés, I.C., Castronuovo, L.,