RESEARCH AND PRACTICE

Digital Junk: Food and Beverage Marketing on Facebook Becky Freeman, PhD, Bridget Kelly, PhD, Louise Baur, PhD, Kathy Chapman, MNutrDiet, Simon Chapman, PhD, Tim Gill, PhD, and Lesley King, MPsych

Obesity is a common, serious, and costly health issue.1 In the United States alone, the medical costs of obesity are estimated to be $147 billion.2 Although the prevalence of overweight and obesity among children and adolescents in countries such as the United States and Australia appears to have plateaued in recent years, rates remain high.3 Obesity rates generally increase with age among adults4; however, of urgent concern is the growing prevalence of overweight and obesity among young Australian adults, particularly women. Research has suggested that later generations have higher rates of excess body weight than generations before them. Current obesitypromoting environments likely mean people are now put at greater risk for weight gain in young adulthood.4 One of the powerful environmental factors influencing the rise in obesity is the ubiquitous presence of food and beverage marketing.5---8 Research into the nature and extent of this marketing has primarily focused on television advertising.9,10 Although there is emerging research on how energy-dense, nutrient-poor (EDNP) food and beverages are being marketed in digital media,11---15 little of this research has closely examined online social media channels.14 Additionally, most of this research on digital media food marketing has focused on Web sites targeted at children and has not captured what types of food marketing adolescents and young adults are most likely to view. Given the exponential growth in popularity of social media Web sites such as Facebook, particularly among adolescents and young adults, there is a need to understand the techniques and reach of EDNP food and beverage marketing on these Web sites. Equally, although case studies of specific campaigns and food companies help to highlight the importance of social media in the marketing mix,16 a more complete picture of overall EDNP marketing strategies used through social media is needed to understand the extent of marketing across this media.

Objectives. We assessed the amount, reach, and nature of energy-dense, nutrientpoor (EDNP) food and beverage marketing on Facebook. Methods. We conducted a content analysis of the marketing techniques used by the 27 most popular food and beverage brand Facebook pages in Australia. We coded content across 19 marketing categories; data were collected from the day each page launched (mean = 3.65 years of activity per page). Results. We analyzed 13 international pages and 14 Australian-based brand pages; 4 brands (Subway, Coca-Cola, Slurpee, Maltesers) had both national and international pages. Pages widely used marketing features unique to social media that increase consumer interaction and engagement. Common techniques were competitions based on user-generated content, interactive games, and apps. Four pages included apps that allowed followers to place an order directly through Facebook. Adolescent and young adult Facebook users appeared most receptive to engaging with this content. Conclusions. By using the interactive and social aspects of Facebook to market products, EDNP food brands capitalize on users’ social networks and magnify the reach and personal relevance of their marketing messages. (Am J Public Health. 2014;104: e56–e64. doi:10.2105/AJPH.2014.302167)

Facebook is the most popular social networking site in the world. As of September 30, 2013, 1.19 billion users accessed the site at least monthly and 727 million users accessed the site daily.17 Approximately 80% of the daily active users are outside the United States and Canada. Australians are enthusiastic Facebook users, with 9 million people, or nearly 40% of the entire population, visiting the site every day.18 Social media use has reached near saturation among young Australians, with more than 85% of those aged 15 to 24 years accessing the Internet for social networking or online gaming.19 The bulk of Facebook content is individual users’ personal profiles, but since November 2007, the site has embraced companies and brands developing their own pages.20 Facebook brand pages function in a similar fashion to personal pages, with the exception that to receive brand page updates and content in their news feed, users must “like” a brand page, as opposed to initiating a friend request as they would from individual users. Brands can post images, videos, links, contests, offers, applications, polls, quizzes, and a range of other digital and interactive media to their page timelines. Users who like brand pages can engage with

e56 | Research and Practice | Peer Reviewed | Freeman et al.

the page by sharing their own content, commenting on page posts, and sharing page content with their own networks.21 Crucially, any activity that users engage with on brand pages may then appear in the news feed of the users’ friends, effortlessly spreading marketing messages across social networks. Facebook uses the EdgeRank algorithm to decide what content will appear in the news feed.22 Generally speaking, those brands with more engaging content have greater success in appearing higher up and more often in news feeds. Readers who are unfamiliar with the layout and function of Facebook pages should refer to online user guides23 and previously published work.24 Definitions of the Facebook terms used in this article are included in Table 1. Given that marketing influences food preferences, choice, and consumption,7 understanding how food is being promoted on social media is essential. The primary aim of this study was to assess the amount, reach, and nature of EDNP food marketing to Australians through Facebook. To begin to build a complete picture of the food and beverage marketing techniques being used on Facebook, we need to know the food and beverage brands that are most active on Facebook, how these

American Journal of Public Health | December 2014, Vol 104, No. 12

RESEARCH AND PRACTICE

TABLE 1—Definitions of Facebook Terms and Marketing Methods Facebook Terms Talking about

Definition The number of unique users who have created a story about a page in a 7-day period. On Facebook, stories are items that display in a news feed. Users create stories when they like a page, post on a page wall, like a post, comment on a post, share a post, answer a question, RSVP to a page’s event, mention the page in a post, tag the page in a photo, check in at a place, share a check-in deal, like a check-in deal, write a recommendation, or claim an offer. This figure is updated daily. (http://www.insidefacebook. com/2012/01/10/people-talking-about-this-defined/)

Most popular age group Timeline

Data publicly available through the Facebook page; a measure of which age group the page is most popular among A sequential (by date) summary of all the activity posted on the Facebook page. It is possible to scroll back through the timeline

News feed

A summary of the activity of the user’s Facebook network that appears when a user logs into the site

until the date on which the page was launched. Marketing category (n = 19) Competitions, prizes, giveaways

Any contest involving a participant entry, including minimal requirements such as simply liking a post; giveaways also include free

Special price promotions

Limited-time offers, discount menus, 2 for 1 deals, or other reduced-price advertisements

Vouchers, offers, rebates

Includes those that consumers print off or for which they enter an electronic code; offers are specific to Facebook and made exclusively available to those who like the page.

product samples and other items with purchase.

Celebrities

People with an entertainment or media profile, excluding athletes.

Children’s characters

Third-party cartoons and characters, including characters from films, books, television programs, and the Internet

Sportspeople

Any person (adult or child) profiled for their athletic or sporting achievements

Branded characters

Any characters featured on the page developed by the brand

Branding elements

Any logos, colors, trademarks, or slogans

Games

Interactive and entertaining applications that feature the brand

Photos Quizzes and polls

Digital images of the product, users, and promotional events Can be embedded directly into the Facebook timeline; they are a feature available to all brand pages as a way of encouraging

Videos

Can either be posted directly to Facebook or linked through YouTube

Events

Specific Facebook category in which page owners can create events and invite page members

Apps

Both links to any smartphone apps and any apps embedded in the Facebook page. Facebook allows page administrators to develop a

participation and interaction.

variety of application tabs on their pages, including retail store location finders, other social media channel feeds, ordering platforms, feedback, and promotional offers. Conversations Links

The page administrator responds to page member posts and comments and shares member content with other members. Any page posts that include a link to an external page or additional content not found within the Facebook page

User-generated content

Digital media (such as photos, videos, songs) created by users and either shared on the page timeline or tagged with the brand

Sponsorships and partnerships

Any events that the brand supports or other brands or services the brand partners with, excluding charitable organizations

Corporate social responsibility

Promotion of any ethical or sustainable initiative or charitable work undertaken by the brand

page by Facebook users

and philanthropy

brands promote their products in terms of the advertising techniques that are used, who is engaging with these brands, and how they engage. We also discuss the potential policy and practice implications of our findings.

METHODS We used a multistep approach to collect and analyze the marketing content of popular food and beverage brand Facebook pages.

Sample Selection On February 1, 2013, we ranked the top 250 Facebook pages on the basis of the total number of Australian Facebook users who had liked the page, using data from the social media monitoring site Socialbakers.25 This list was freely available, with all data sourced directly from Facebook. This list was not limited to Australian-based brand pages but also included international Facebook pages popular among Australian users.

December 2014, Vol 104, No. 12 | American Journal of Public Health

To determine our final sample within these 250 pages, we excluded any pages that did not represent food and beverage products or companies and included food manufacturers, food brands, retailers, and food service providers such as restaurants. We excluded alcohol brands because these Facebook pages only allow members who state they are of legal purchase age to join. After this process, 33 pages that had a connection to food and beverage brands remained. Of these 33 pages, 3 were

Freeman et al. | Peer Reviewed | Research and Practice | e57

RESEARCH AND PRACTICE

Australian food retailers. We subsequently excluded these pages because an initial examination of the content of their Facebook pages showed that they promoted non---food-related products and services (e.g., cleaning products). We eliminated 2 other pages from the final sample because these pages were genuine fan communities and were not professionally moderated. Finally, we removed the international brand page for Pringles potato chips from the sample because Australian Facebook users are now automatically redirected to the Australiafocused page, which was already captured in the study sample. Our final sample consisted of 27 food and beverage brand pages.

Data Collection We recorded the Facebook page names, the total number of Australian Facebook likes the page had received, the total number of global Facebook likes the page had received (including from Australia), and the page’s rank. Content included in the analysis was taken from all posts made by the page to its Facebook timeline from each page’s launch date through February 19, 2013. We developed a content analysis coding tool to categorize the marketing techniques used by the brands. We assessed each piece of content posted with respect to what type of marketing technique was being used; for example, was a celebrity included, were users being offered an incentive or prize, or was the company promoting its own socially responsible behavior? Initially, we included 17 different categories of possible marketing techniques in the tool, based on previous content analyses of EDNP food and beverage television marketing14,26 and case studies of social media marketing tactics.27 This tool was piloted on a globally popular food brand Facebook page that was not part of our sample. Two of us (B. F. and B. K.) jointly collected and coded the content to assess its appropriateness and completeness and determine which data would be collected from the page. After this pilot, we added an additional 2 categories to the coding tool. These 19 categories of marketing techniques are described in Table 1 and include, for example, the use of celebrities, vouchers, games, competitions, apps, user-generated content (UGC), and price promotions. In addition, we collected key page

statistics, including the date the page was launched, the number of people talking about the page, the most popular age group to like the page, and whether consumers could post directly to the page timeline. We scrolled through the posts-by-page timeline and noted any examples of the 19 marketing techniques in an Excel sheet developed for each page. We chose the posts-by-page filter because this content is what users who like the page are most likely to see. Also, posts by others are not shared in the default timeline of the page and, when examined in the pilot, were mostly made up of complaints about or accolades for the brand posted by users that had attracted very little comment. Each page had an average of 3.65 years of activity. We calculated descriptive statistics for each category and identified illustrative examples. For 3 of the 19 marketing techniques (competitions, prizes, and giveaways; games; apps), we conducted a detailed content analysis of the page posts that related to these categories. We selected these 3 categories for more detailed analysis because of the variety and innovation we noted when collecting the data and the unique marketing approaches that social media allow. We have included case examples from the brand pages to illustrate these marketing methods. Last, we collected information from the pages to estimate the reach of the marketing messages posted. This information was collected for all original posts made by the page administrators to the timeline only over a 1-month period between January 19 and February 19, 2013, including the number and type of posts the page had made and the number of Facebook users who had liked, commented on, or shared each of these individual posts. One estimate of how engaged page members are beyond initially liking the page is to assess the people talking about versus the total number of likes—this gives the percentage of the page membership that is actively engaged with the page over the preceding 7-day period. When people interact with pages in ways that generate news feed stories, pages reach an audience beyond their existing fan base, which is essential to generating additional word-of-mouth promotion.28 We calculated the percentage of people talking about the page as a proportion of the total number of likes for each page.

e58 | Research and Practice | Peer Reviewed | Freeman et al.

RESULTS All of the food and beverage brand pages in the final sample were classified as EDNP dietary choices. EDNP food and beverages are high in calories but low in nutritional value. Although some of the fast food restaurants included in the sample do offer healthy options, observational studies have indicated that regular (less healthy) menu items make up the bulk of food sales in these outlets.29 Of the 27 pages, 7 were for fast food restaurants, 5 for chocolate, 4 for sugar-sweetened sodas, 3 for energy drinks, 2 for confectionery brands, 2 for ice cream brands, 2 for condiments or spreads, and 1 each for a sweet biscuit and a salty snack. Table 2 summarizes the content of each page. There was a nearly even mix of international (13 pages) and Australian-based brand pages (14 pages), with 4 brands (Subway, Coca-Cola, Slurpee, Maltesers) represented by both an international and an Australian version of the page. The most popular Facebook food and beverage brand page in Australia was Bubble O’Bill Ice Cream, with a total of 1 055 065 Australian likes and a global page membership of 1 281 834. Bubble O’Bill Ice Cream is an Australian product produced by the Streets company. By comparison, the least popular page in the list, Maltesers Australia, had 271 389 Australian likes and a total of 281 480 global likes. Overall, across the 27 pages, there were a total of 13 191 041 Australian likes and 277 548 679 global likes. In terms of age groups, pages were most commonly liked by those aged 18 to 24 years, with 16 of the pages most frequently liked by this age group. Five pages (Maltesers Australia, Cold Rock Ice Creamery, Slurpee Australia, Subway Australia, Coca-Cola Australia) were most popular among those aged 13 to 17 years. In addition, those aged 13 to 24 years were the most common age group to like 4 pages: Domino’s Pizza Australia, Pringles, McDonald’s Australia, and Cadbury Eyebrows. The remaining 2 pages, Vegemite (a savory, salty spread popular in Australia) and Cadbury Dairy Milk Australia, were most popular with a slightly older audience, those aged 25 to 34 years. All pages were professionally moderated and appeared to be administered by either the company brand owner or an authorized advertising agency. All but 5 pages allowed

American Journal of Public Health | December 2014, Vol 104, No. 12

RESEARCH AND PRACTICE

TABLE 2—Summary Description of the Facebook Pages as of February 1, 2013 Name Bubble O’Bill Ice Creams

Ranka 9

Category of Brand

Date Page Launched

Most Popular Age Group, Years

Australian Likes, No.

Global Likes (Includes Australia), No.

Talking About, No.

Members Talking About, %

Ice cream

July 29, 2008

18–24

1 055 065

1 281 834

5102

Skittles

16

Confectionary

December 3, 2007

18–24

896 469

24 435 851

181 425

0.4 0.7

Domino’s Pizza—Australia

24

Quick-serve restaurant

January 4, 2009

13–24

762 990

822 714

5169

0.6

Coca-Cola Australia Coca-Colab

25 30

Sugar-sweetened beverage Sugar-sweetened beverage

December 18, 2008 December 15, 2008

13–17 18–24

761 020 714 465

922 943 59 576 447

4509 1 430 168

0.5 2.4

Red Bull

34

Energy drink

November 15, 2007

18–24

689 609

36 142 113

407 760

1.1

Pringles (Australia region)

40

Salty snack

July 27, 2009

13–24

664 532

22 650 276

17 761

0.1

Oreo

54

Baked snack

August 7, 2009

18–24

590 894

31 889 266

198 526

0.6

KFC Australia

65

Quick-serve restaurant

June 12, 2009

18–24

542 813

654 685

4246

0.6

McDonald’s Australia

79

Quick-serve restaurant

February 17, 2011

13–24

496 615

553 761

12 676

2.3

Subway Australia

97

Quick-serve restaurant

June 1, 2010

13–17

452 133

492 107

14 727

3.0

Slurpee Monster Energy

122 131

Sugar-sweetened beverage Energy drink

June 26, 2008 April 20, 2008

18–24 18–24

416 365 409 940

4 884 844 21 073 042

57 839 296 637

1.2 1.4

Ferrero Rocher

132

Chocolate

March 15, 2008

18–24

409 314

17 414 997

30 548

0.2

Subway

133

Quick-serve restaurant

December 16, 2009

18–24

401 082

20 582 657

158 980

0.8

Starburst

135

Confectionary

September 10, 2009

18–24

400 084

11 876 839

29 557

0.2

Pizza Hut Australia

148

Quick-serve restaurant

November 16, 2010

18–24

383 920

413 894

1801

0.4

Slurpee Australia

149

Sugar-sweetened beverage

January 20, 2010

13–17

383 618

406 706

4726

1.2

V Energy Drink Australia

155

Energy drink

September 26, 2010

18–24

377 761

499 664

11 634

2.3

Maltesers Nutella

165 207

Chocolate Spread

May 29, 2012 July 28, 2008

18–24 18–24

357 736 309 651

1 625 219 17 086 900

14 754 64 647

0.9 0.4

Cold Rock Ice Creamery

210

Ice cream

February 10, 2008

13–17

305 780

330 822

129

0.0

Cadbury Dairy Milk—Australiac

214

Chocolate

March 22, 2010

25–34

298 784

381 288

31 344

8.2

Cadbury Eyebrows

228

Chocolate

January 23, 2009

13–24

284 819

626 159

157

0.0

Vegemite

234

Spread

July 1, 2009

25–34

280 469

329 526

4415

1.3

Hungry Jack’s

244

Quick-serve restaurant

May 20, 2008

18–24

273 724

312 645

16 212

5.2

Maltesers Australia

250

Chocolate

January 15, 2012

13–17

271 389

281 480

10 491

3.7

Totals Average

13 191 041 488 557

277 548 679 10 279 581

3 015 940 111 701

1.1 1.5

Median

13 191 041

922 943

14 754

0.8

Note. Duplication in the fan base across pages is likely, but we were unable to quantify this. a Rank is within Top 250 Facebook Pages in Australia as of February 1, 2013. b Unable to open timeline photo album. c At the time of data collection, this was a stand-alone Facebook page. It is now part of the Cadbury Dairy Milk page, but the same page timeline content appears in the Australia region filter.

Facebook page members to post their own content to the brand-page timeline (referred to as “posts by others”). However, all pages allowed users to respond to and share content that the page had posted. Most pages had a very small percentage of members actively engaged in the week preceding data collection, with a median of 0.77% of members talking about the page. Coca-Cola had the most engagement in terms of absolute numbers, with 1 430 168 talking about the page (2.4% of total likes). Cadbury Dairy Milk Australia had the leading amount of talking-about engagement in terms

of a percentage of total likes, with 8.2% (n = 31 344).

Marketing Techniques Table 3 presents an overview of the types of marketing techniques used by the pages. All pages included photographs and branding elements, such as logos, trademarks, slogans, and brand colors. Of the 27 pages, 17 highlighted their philanthropy or corporate social responsibility activities and 21 included content that showcased their sponsorship of events or partnerships with other brands, companies,

December 2014, Vol 104, No. 12 | American Journal of Public Health

or services. Table A (available as a supplement to this article at http://www.ajph.org) provides further detail about the pages that used these techniques. The majority of brand pages used celebrities, licensed characters popular with children, and sportspeople to promote their products. Nearly half of all pages (n = 13) included a branded character created by the company to promote its products. Although most pages that included these high-profile people and licensed characters had an obvious paid relationship, this was not universally the case. For example, the

Freeman et al. | Peer Reviewed | Research and Practice | e59

RESEARCH AND PRACTICE

TABLE 3—Marketing Techniques Used by the Facebook Pages as of February 1, 2013 Category Branding elements

Pages, No. (%) 27 (100)

Photos

27 (100)

User-generated content

25 (92.6)

Competitions, prizes, and giveaways

24 (88.9)

Apps Videos

24 (88.9) 23 (85.2)

Conversations

23 (85.2)

Links

23 (85.2)

Allow posts by others

22 (81.5)

Celebrities

22 (81.5)

Sponsorships or partnerships

21 (77.8)

Children’s characters

19 (70.4)

Quizzes or polls Sportspeople

19 (70.4) 17 (63.0)

Corporate social responsibility

17 (63.0)

or philanthropy Branded characters

13 (48.1)

Vouchers, offers, and rebates

12 (44.4)

Events

11 (40.7)

Games

9 (33.3)

Special price promotions

8 (29.6)

Maltesers Australia page focused on eating Maltesers while watching movies either in the theater or at home and had posted several trivia questions about, and quotes from, films. The page also regularly posted a “Who would you rather share a pack of Maltesers with?” image that included a photo of 2 celebrities and asked Facebook followers to select their preference. It is unlikely that Maltesers has any kind of formal relationship with these actors, yet images of these people are shown, and Maltesers is able to connect its product to famous names without any formal contract. Similarly, with characters popular with children, some pages posted images of popular toys and cartoons and asked followers to nominate their childhood favorite. Other pages had more traditional marketing product tie-ins, with Hungry Jack’s and Subway, for example, offering purchase deals that included a toy with a children’s meal. Sportspeople, in particular Australian Olympic athletes, were repeatedly mentioned on brand pages when they had won medals at the London 2012 Olympics. Again, although some brands—Coca-Cola,

Cadbury, and McDonalds, for example—were official corporate sponsors of the Olympic games, no indication was made of any formal relationship on many of the other pages that discussed the games and athlete achievements. It is entirely possible that these associations were a deliberate form of ambush marketing— the unauthorized association with an event by an advertiser with a view toward exploiting its goodwill for commercial purposes.30 All 7 fast food restaurant pages posted special price promotions on their pages. The only other brand to share a special price promotion with Facebook members was the Slurpee Australia page. Slurpees are frozen sugar-sweetened beverages that can be purchased at 7-Eleven convenience stores. All fast food restaurants offered a voucher, offer, or rebate to their Facebook followers. These vouchers were often exclusive to Facebook members and available for a limited time or number. Five other pages—Slurpee, Slurpee Australia, Cadbury Dairy Milk Australia, Cold Rock Ice Creamery, and Oreo—also offered these sorts of offers to their followers. More than 70% of pages included videos, quizzes or polls, links, and apps as part of their social media marketing efforts. A smaller number of pages included games (n = 9 pages) or made use of the Facebook events tool (n = 11 pages). The use of features unique to social media that increased consumer interaction and engagement was widespread across all pages. All but 2 pages (Ferrero Rocher and Cadbury Eyebrows) featured UGC. A typical example of UGC is a consumer taking a photo of the product or of themselves eating or drinking the product and tagging the brand page in the photo. The page can then in turn share the photo with all the followers on the page so that the image reaches well beyond the social network of the user. Twenty-three pages contained evidence of conversations between the brand page and Facebook followers, with the page responding directly to user posts, liking and replying to comments made by users, or sharing user posts. Bubble O’Bill Ice Cream was exceptional in that the page responded to virtually every post made by others to its timeline and engaged with post comments daily, which may explain why it was the most popular page.

e60 | Research and Practice | Peer Reviewed | Freeman et al.

Detailed Content Analysis Competitions, prizes, and giveaways. Competitions, giveaways, or prizes were a frequent marketing technique and were evident on 24 of the 27 pages, with only Cadbury Eyebrows, Nutella, and Ferrero Rocher not offering these incentives. Seventeen pages included links to, or promotions of, contests that were being conducted offline or on sites outside of Facebook. Fourteen pages had held a UGC competition in which followers submitted ideas, photographs, videos, songs, letters, or poems in exchange for being entered in the competition. Often, winners received no material prize for UGC; instead, the winner would be featured on the brand page as a fan of the week or biggest fan. For example, Skittles (confectionery) capitalized on user-generated photos of fans eating or using Skittles in amusing ways as part of their Greatest Fan in the World series of posts. Five pages had simple giveaways in which followers were automatically entered in the contest if they liked or commented on a page post or if the page reached a certain number of likes, at which stage a special deal would be offered to users. Typical giveaways included product samples or free side orders or desserts at fast food restaurants. Thirteen pages had contests or giveaways that were exclusively available for Facebook members, mobile app users, or other social media followers (such as Twitter and Instagram). Nine pages held contests that selected a winner by popular vote of page members. Games. Nine of the brand pages included a digital game. Games on the Nutella (hazelnut chocolate spread) and Coke Australia pages were very simple word searches and word riddles featuring product attributes embedded in the Facebook timeline. The Domino’s Pizza Australia game was a photo-based flip book that followed a cartoon pizza delivery scooter. In contrast, Red Bull included a Facebook page tab that linked to an external site with more than a dozen games, both computer and mobile phone accessible, based on the motor sports that Red Bull sponsors. Coca-Cola similarly had links to an external game called Crabs and Penguins that can be downloaded to mobile phone or tablet devices. In a promotional YouTube video (watched by 1.7 million people), Coca-Cola says the game is aimed at

American Journal of Public Health | December 2014, Vol 104, No. 12

RESEARCH AND PRACTICE

people aged 13 years and older,31 despite its featuring animated animals and simple gaming tasks that could be achieved by younger children. Online reviewers of the game were impressed with the high-quality graphics and game play, despite acknowledging it was essentially an advertisement for Coke.32 Slurpee launched an online game, Rock, Paper, Scissors, as part of the promotion for a sports drink---based Slurpee. The game was hosted on an external site and heavily promoted on the Facebook page. Players were eligible to win free Slurpees and other prizes such as chocolate bars and T-shirts. McDonald’s had a simple Facebook digital game of running with a container of fries across the beach while avoiding seagulls who try and steal the fries. Beating one’s best score and those of other players encouraged repeat plays. Slurpee Australia, Red Bull, and Oreo had games embedded in mobile phone apps promoted on their pages. The Cadbury Facebook game, Cadbury’s Charm Academy, was no longer available for viewing but had previously been promoted on the Cadbury Eyebrows campaign page. Apps. Twenty-four pages included links to smartphone or Facebook page apps in their page timelines. Seven of the pages included a link to a smartphone app that could be downloaded from either iTunes or Google Play stores. These smartphone apps had a range of functions, such as finding movie session times and cinema locations (Maltesers Australia), playing games (Red Bull, Slurpee Australia, Oreo), tracking the ingredients and nutrition of some menu items (McDonald’s), and redeeming vouchers and claiming prizes (Hungry Jack’s). Red Bull also had several apps to accompany the variety of sponsored extreme and motor sport events. Vegemite developed a “wake up and smell the Vegemite” smartphone app to coincide with the London 2012 Olympics so that users could set alarms to avoid missing their favorite events; the product marketing premise was presumably that users would also make toast and Vegemite at 4:00 A.M. while watching Australian athletes. Vegemite partnered with the Australian swimming coach and an Australian comedian to act as the voices of the App alarms. Vegemite developed a Facebook app to coincide with the launch of its My First Vegemite product, a lower salt variety of the original product aimed at children (the product is still classified

as high in salt).33 Called “Magical Moments,” the app was a tool for parents to track their children’s growth and milestones, including tasting the new Vegemite. Four pages—Subway Australia, Domino’s Australia, Pizza Hut Australia, and Maltesers— included Facebook apps that allowed page followers to place an order without having to click outside of Facebook. Domino’s promoted its Facebook-ordering app by offering exclusive prices and menu upgrades to its Facebook customers. Other Facebook apps included store or restaurant location finders; feedback surveys; offers; news feeds from other social media platforms such as Twitter, YouTube, and Instagram; and competitions.

Summary of 1 Month of Page Activity Across the pages, there was an enormous range of activity over the month-long period of analysis (Table 4 provides a complete summary). Some page administrators posted content multiple times per day, whereas other page administrators posted no new content for the entire month. The majority of posts across all the pages were of photographs, in which even purely text-based posts were presented as photographs by stylizing the text as an image file. On average, page administrators made a total of 18 original posts during the 1-month period, of which 13 (72%) were classified as photographs. The most active page was the Monster Energy drink page, with 67 posts in 1 month. All page posts attracted likes, shares, and comments from page members. Monster Energy had the highest total number of likes for its posts across the 1-month period with 1 281 868 total likes, and Subway had the highest average number of likes per post with 23 569 likes. Red Bull had the highest average of number of shares per post with 1334 shares, and Starburst attracted the highest average number of comments per post with 3960 comments. Given that a significant portion of Facebook users log in daily, it is unsurprising that popular pages have high levels of activity. Super Bowl XLVII (February 3), Valentine’s Day (February 14), and Australia Day (January 26) fell within the month of data collection. Most pages posted a Valentine’s Day---themed photograph, mentioned the Super Bowl in a post, and shared wishes for a Happy Australia Day, including the international brand page

December 2014, Vol 104, No. 12 | American Journal of Public Health

for Skittles. All Australian brand pages posted patriotic Australia Day messages that tied being Australian to the purchase and consumption of their product. For example, KFC posted an image of the Australian coat of arms in which the central shield was replaced by a bucket of KFC chicken (Figure 1). The image was also used to promote a KFC-sponsored performance by musicians the Madden brothers at the Australia Day cricket match that was being aired on television that evening. The post attracted 4069 likes, 380 shares, and 233 comments from page members.

DISCUSSION It is clear from our study that EDNP brands are using the interactive and social aspects of Facebook to market their products. Adolescent and young adult Facebook users appeared equally receptive to engaging with this content, as evidenced by this group being the most frequent to like these pages. We should also note that there is no way to verify the age of users, an issue Facebook itself has acknowledged, and many users may be younger than 13 years.34 These consumers not only willingly engage with brands, they also create free word-of-mouth content that marketers have minimal control over. This kind of consumer involvement and engagement is unique to social media communication. Although we found that only a small percentage of the total page members were actively engaged with the pages, given the high number of total users this audience size is significant. Although not captured in the page statistics of likes and shares, this audience can also extend to the friends list of users who interact with the brand page by appearing as content in their friends’ news feed. Additionally, these statistics do not account for the large number of users who may have seen some page content in their news feed but did not elect to interact with it in any way. Exposure to these marketing messages potentially goes well beyond the smaller group of highly engaged Facebook users. The reach of social media marketing messages should be assessed not just in terms of audience engagement but also in total audience exposure. With pages posting new content on average approximately every 2 days, and some pages posting multiple times a day, combined with the daily log-in habits of Facebook

Freeman et al. | Peer Reviewed | Research and Practice | e61

RESEARCH AND PRACTICE

TABLE 4—Timeline Posts by Page: January 19–February 19, 2013 Name

Posts by Page Only, No.

Photograph-Based, No. (%)

Likes (Total for All Posts), No.

Shares (Total for All Posts), No.

Comments (Total for All Posts), No.

Mean Likes per Post

Mean Shares per Post

Mean Comments per Post

Subway

30

24 (80.0)

707 083

25 843

28 100

23 569

861

937

Red Bull

26

23 (88.5)

522 052

34 680

5445

20 079

1334

209

Monster Energy

67

62 (92.5)

1 281 868

77 635

13 220

19 132

1159

197

Starburst Nutella

2 10

1 (50) 10 (100)

16 115 76 237

103 7437

7919 2020

8058 7624

52 744

3960 202

Oreo

6

5 (83.3)

44 836

2888

872

7473

481

145

Maltesers Australia

7

7 (100)

26 756

2710

625

3822

387

89

Bubble O’Bill Ice Creams

9

7 (77.8)

27 978

801

551

3109

89

61

Cadbury Dairy Milk Australia

30

22 (73.3)

67 596

6318

7104

2253

211

237

Maltesers

14

12 (85.75)

23 652

2033

515

1689

145

37

Skittles

23

9 (39.1)

34 365

1233

1031

1494

54

45

McDonald’s Australia Pringles (Australia region)

14 32

4 (28.6) 32 (100)

20 169 40 787

3265 719

5210 1476

1441 1275

233 22

372 46

Domino’s Pizza—Australia

10

3 (33.3)

12 285

624

4139

1229

62

414

Vegemite

20

11 (55.0)

18 246

4752

3472

912

238

174

Slurpee Australia

25

25 (100)

17 721

288

554

709

12

22

KFC Australia

28

25 (89.3)

19 758

1118

3066

706

40

110

Subway Australia

7

3 (42.9)

2960

133

357

423

19

51

V Energy Drink Australia

23

12 (52.2)

8763

275

2420

381

12

105

Coca-Cola Australia Hungry Jack’s

14 33

8 (57.1) 21 (63.6)

4831 10 709

288 756

712 1234

345 325

21 23

51 37

Slurpee Pizza Hut Australia

1 27

0 (0) 12 (44.4)

151

11

34

151

11

34

3639

141

1192

135

5

44

Coca-Colaa

0

0

0

0

0

0

0

0

Ferrero Rocher

0

0

0

0

0

0

0

0

Cold Rock Ice Creamery

0

0

0

Average

17.6

13 (73.7)

Median

14.0

9.5 (53.9)

114 945 19 002.0

0

0

0

6694.3

0

3510.3

0

4090

239

292

778.5

1213.0

1252

53

75

Note. The Cadbury Eyebrows page was not included because a final announcement on the page said no more posts would be made as of December 23, 2012, because the campaign had concluded. a Unable to open timeline photo album; timeline had no posts for the time period.

users, the reach of marketing messages quickly amplifies. Despite some early research that suggested that users were opposed to engaging with large corporations on Facebook unless they received significant rewards,35 our study shows that users require very little incentive to openly interact with EDNP food brands. The use of competitions and giveaways as a persuasive form of advertising in EDNP food is well established.36 Social media build on the effectiveness of competitions by allowing companies and consumers to collaborate and build brands together.37 Facebook provides unique ways for consumers to enter contests that at the same time promote the brand among the users’ social network. The prizes offered by the brands in

our study were generally of low monetary value, but some did provide users with the opportunity to be a featured page member or fan of the week. This suggests that increasing the visibility of users on social media among their peers—or fellow consumers—is a distinctive social media marketing tactic. The type of food and beverage pages popular on Facebook is reflective of the food choices of Australian adolescents38 and young adults.39 Of particular note is the popularity of the sugar-sweetened soda and energy drink pages. Increased consumption of these beverages is an established contributor to rising levels of obesity40 and has been the focus of controversial tax policy reform. The promotion and availability of EDNP food and

e62 | Research and Practice | Peer Reviewed | Freeman et al.

beverages is ubiquitous in the offline environments of young Australians and, as we have documented, has a similarly dominant presence in their online social media networks. The pages that encourage and enable consumers to actually purchase products within the Facebook page serve as a further cue for purchase and consumption of these EDNP foods.

Public Health Practice Implications In terms of health policy, much of the current work to limit exposure to EDNP advertising is focused on restricting advertisements during children’s television programs and viewing hours. Our study shows that this narrow focus is likely to miss large amounts of online advertising aimed at adolescents.

American Journal of Public Health | December 2014, Vol 104, No. 12

RESEARCH AND PRACTICE

Note. The post included the caption “Happy Straya Day! Tune into Channel 9 tonight to watch the Maddens perform live at the cricket at half time! C’mon you Aussies!” (Straya is local slang for Australia.)

FIGURE 1—KFC Australia Day Facebook brand page post: January 26, 2013.

Additionally, young adults appear to be a highly desirable target population for EDNP food marketing, and limited research, resources, and policy action have been directed at this age group. Emerging adulthood (ages 18---25 years) is largely overlooked for establishing long-term health behavior patterns. Factors such as identity development and shifting interpersonal influences differentiate young adulthood from other life stages and are likely to influence the adoption of both healthy and unhealthy behaviors.41 As a minimal first step, increased monitoring of how EDNP food and beverages are marketed on social media is essential. Although our study focused on Australian Facebook users, our findings have international relevance given that many of the pages in our study were for global brands. Previous research has found that people who experienced strong positive emotions while viewing Facebook page content for food and beverage brands were 3.25 times more likely to claim that they would recommend the brands and 2.5 times more likely to prefer the brands.42 It is unsurprising, then, that the Facebook pages in our study were quick to align themselves with positive, fun events such as Australia Day and the Super Bowl. Conversely, some of the most effective public health behavior change campaigns have generated high negative emotional arousal and

may not be as effective in social media environments in which people can actively avoid these uncomfortable messages.43 If people are engaging with Facebook content because it makes them feel good, it may mean that certain modes of health promotion messages that are highly effective in other forms of media will not work on social media. Our study shows that large numbers of people are willing to allow EDNP corporations into the same social network as their friends, relatives, and so forth and that they are either unaware or unconcerned that engaging with these companies generates enormous amounts of virtually free marketing content. Public health agencies could capitalize on this same effect by dedicating resources to building strong Facebook communities. The Hello Sunday Morning Facebook page, which encourages followers to take a break from using alcohol, is an example of how users can be engaged to direct action and also support others.44

evident in our study when the Cadbury Dairy Milk Australia page merged with the Cadbury Dairy Milk page. Nor was it possible to determine whether the same Facebook users had liked multiple brands’ pages, for example, liking both Coca-Cola and McDonald’s. In addition, both the page content and how that content is displayed in Facebook news feeds changes regularly. Pages and specific pieces of content can easily climb up and fall down the social media ladder. For this study, we included only unpaid Facebook marketing methods. Pages can also market their products through paid advertising by paying to preferentially promote page posts to existing members and to place ads either directly in the news feeds or on the side bars of any Facebook user’s news feed. This study is part of a larger grant investigating the nature, reach, and impact of EDNP food and beverage marketing on social media. A great deal of potential exists to further investigate how engagement and interaction are stimulated on these pages. Potential future work includes in-depth analysis of the types of UGC that attracted the most attention from other users and the nature of conversations between Facebook users and the brand owners. Our next research project is a survey of young social media users that will explore their views of EDNP food and beverage marketing on Facebook and associated behaviors, particularly if they have created UGC. This is the first study to our knowledge to systematically assess the nature of food promotions on the globally popular social media site Facebook. Study findings show that EDNP food and beverage marketing is prolific and seamlessly integrated within online social networks. Adolescents and young adults are engaging with EDNP brands on a near-daily basis. Facebook users willingly spread marketing messages on behalf of food and beverage corporations with seemingly little incentive or reward required. j

Limitations It was not possible to determine, using publicly available data, how many of the fans liked both the Australian and the international page versions of the brand, although Facebook has introduced filters that allow brands to channel users to country-specific pages and content on the basis of their profile. This was

December 2014, Vol 104, No. 12 | American Journal of Public Health

About the Authors Becky Freeman, Simon Chapman, and Lesley King are with the School of Public Health, University of Sydney, New South Wales, Australia. Bridget Kelly is with the School of Health and Society, University of Wollongong, New South Wales. Louise Baur is with the Discipline of Paediatrics and Child Health, University of Sydney, New South Wales. Kathy Chapman is with Health Strategies, Cancer Council New South Wales, Woolloomooloo, Australia. Tim Gill is

Freeman et al. | Peer Reviewed | Research and Practice | e63

RESEARCH AND PRACTICE

with the Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney, New South Wales. Correspondence should be sent to Becky Freeman, School of Public Health, Sydney Medical School, University of Sydney, 226a, Edward Ford Building (A27), Sydney, New South Wales 2006, Australia (e-mail: becky.freeman@ sydney.edu.au). Reprints can be ordered at http://www.ajph. org by clicking the “Reprints” link. This article was accepted June 27, 2014.

Contributors B. Freeman conceptualized the study, conducted the analysis, and wrote the first draft of the article. B. Freeman and B. Kelly developed the coding tool and collected the data. B. Kelly oversaw the pilot data collection. All authors contributed to the design of the study, development of the methods, and the final draft of the article.

Acknowledgments This study was funded by a competitive research grant from the Australian National Preventive Health Agency titled “Online Food and Beverage Marking to Children and Adolescents” (64FRE2011).

10. Powell LM, Szczypka G, Chaloupka FJ. Adolescent exposure to food advertising on television. Am J Prev Med. 2007;33(4 suppl):S251---S256. 11. Ustjanauskas AE, Harris J, Schwartz M. Food and beverage advertising on children’s web sites. Pediatr Obes. 2013;Epub ahead of print. 12. Cheyne AD, Dorfman L, Bukofzer E, Harris JL. Marketing sugary cereals to children in the digital age: a content analysis of 17 child-targeted websites. J Health Commun. 2013;18(5):563---582. 13. Lascu D-N, Manrai AK, Manrai LA, Amissah FB. Online marketing of food products to children: the effects of national consumer policies in high-income countries. Young Consumers. 2013;14(1):19---40. 14. Kelly B, Bochynska K, Kornman K, Chapman K. Internet food marketing on popular children’s websites and food product websites in Australia. Public Health Nutr. 2008;11(11):1180---1187. 15. Moore ES, Rideout VJ. The online marketing of food to children: is it just fun and games? J Public Policy Marketing. 2007;26(2):202---220.

Human participant protection was not required because the study did not involve human participants.

16. Montgomery KC, Grier SA, Chester J, Dorfman L. The digital food marketing landscape: challenges for researchers. In: Williams JD, Pasch KE, Collins CA, eds. Advances in Communication Research to Reduce Childhood Obesity. New York, NY: Springer; 2013:221---242.

References

17. Facebook. Newsroom. 2013. Available at: http:// newsroom.fb.com/Key-Facts. Accessed November 25, 2013.

Human Participant Protection

1. Centers for Disease Control and Prevention. Adult obesity facts. Available at: http://www.cdc.gov/obesity/ data/adult.html. Accessed August 24, 2014. 2. Finkelstein EA, Trogdon JG, Cohen JW, Dietz W. Annual medical spending attributable to obesity: payerand service-specific estimates. Health Aff (Millwood). 2009;28(5):w822---w831. 3. Rokholm B, Baker J, Sørensen T. The levelling off of the obesity epidemic since the year 1999—a review of evidence and perspectives. Obes Rev. 2010;11(12):835---846. 4. Allman-Farinelli MA, Chey T, Bauman AE, Gill T, James WP. Age, period and birth cohort effects on prevalence of overweight and obesity in Australian adults from 1990 to 2000. Eur J Clin Nutr. 2008; 62(7):898---907. 5. Chandon P, Wansink B. Does food marketing need to make us fat? A review and solutions. Nutr Rev. 2012; 70(10):571---593.

18. Godfrey M. Facebook checked by 9 million Australians every day. The Sydney Morning Herald. 2013. Available at: http://www.smh.com.au/digital-life/digital-lifenews/facebook-checked-by-9-million-australians-every-day20130820-2s7wo.html. Accessed November 25, 2013. 19. Australian Bureau of Statistics. Household use of information technology, Australia, 2010-11. 2011. Available at: http://www.abs.gov.au/AUSSTATS/abs@. nsf/DetailsPage/8146.02010-11?OpenDocument. Accessed November 25, 2013. 20. Greenstein H. Facebook pages vs Facebook groups: What’s the difference? 2009. Available at: http:// mashable.com/2009/05/27/facebook-page-vs-group. Accessed November 25, 2013. 21. Weaver J. The evolution of Facebook for brands. 2012. Available at: http://mashable.com/2009/05/27/ facebook-page-vs-group. Accessed November 25, 2013. 22. Edgerank. What is Edgerank? 2014. Available at: http://www.whatisedgerank.com. Accessed May 29, 2014.

6. Zimmerman FJ. Using marketing muscle to sell fat: the rise of obesity in the modern economy. Annu Rev Public Health. 2011;32(1):285---306.

23. Buck S. The beginner’s guide to Facebook. 2012. Available at: http://mashable.com/2012/05/16/ facebook-for-beginners. Accessed November 25, 2013.

7. Cairns G, Angus K, Hastings G, Caraher M. Systematic reviews of the evidence on the nature, extent and effects of food marketing to children: a retrospective summary. Appetite. 2013;62:209---215.

24. Freeman B, Chapman S. British American Tobacco on Facebook: undermining article 13 of the global World Health Organization Framework Convention on Tobacco Control. Tob Control. 2010;19(3):e1---e9.

8. World Health Organization. Diet, nutrition, and the prevention of chronic diseases: report of a joint WHO/ FAO expert consultation. 2005. WHO technical report series 916. Available at: http://whqlibdoc.who.int/trs/ who_trs_916.pdf. Accessed May 29, 2014.

25. Socialbakers. Facebook statistics by country. 2013. Available at: http://www.socialbakers.com/facebookstatistics. Accessed February 1, 2013.

9. Adams J, Hennessy-Priest K, It S, Sheeshka J, Østbye T, White M. Changes in food advertisements during “prime-time” television from 1991 to 2006 in the UK and Canada. Br J Nutr. 2009;102(4):584---593.

26. Hebden L, King L, Kelly B. The art of persuasion: an analysis of techniques used to market foods to children. J Paediatr Child Health. 2011;47(11):776---782. 27. Freeman B, Chapman S. Gone viral? Heard the buzz? A guide for public health practitioners and researchers on how Web 2.0 can subvert advertising

e64 | Research and Practice | Peer Reviewed | Freeman et al.

restrictions and spread health information. J Epidemiol Community Health. 2008;62(9):778---782. 28. Darwell B. “People talking about this” defined. 2012. Available at: http://www.insidefacebook.com/ 2012/01/10/people-talking-about-this-defined. Accessed January 6, 2013. 29. Wellard L, Glasson C, Chapman K. Sales of healthy choices at fast food restaurants in Australia. Health Promot J Austr. 2012;23(1):37---41. 30. James M, Osborn G. London 2012 and the impact of the UK’s Olympic and Paralympic legislation: protecting commerce or preserving culture? Mod Law Rev. 2011;74(3):410---429. 31. Coca-Cola. A free game from Coca-Cola—“Crabs and Penguins.” 2012. Available at: http://www.youtube. com/watch?v=dZhvwGxXLuY. Accessed January 6, 2014. 32. Wittenkeller J. The best commercial you’ll ever play. 2012. Available at: http://www.gamezebo.com/games/ crabs-and-penguins/review. Accessed January 6, 2014. 33. Miletic D. New Vegemite raises ire of health experts. 2011. Available at: http://www.essentialkids.com.au/ younger-kids/kids-nutrition-and-fitness/new-vegemiteraises-ire-of-health-experts-20110220-1b17z.html. Accessed January 9, 2014. 34. Sweney M. Facebook admits it is powerless to stop young users setting up profiles. The Guardian. Available at: http:// www.theguardian.com/technology/2013/jan/23/facebookadmits-powerless-young-users. Accessed May 29, 2014. 35. Vorvoreanu M. Perceptions of corporations on Facebook: an analysis of Facebook social norms. J New Commun Res. 2009;4(1):67---86. 36. Kelly B, Hattersley L, King L, Flood V. Persuasive food marketing to children: use of cartoons and competitions in Sydney commercial television advertisements. Health Promot Int. 2008;23(4):337---344. 37. Freeman B, Chapman S. Open source marketing: Camel cigarette brand marketing in the “Web 2.0” world. Tob Control. 2009;18(3):212---217. 38. Rangan AM, Randall D, Hector DJ, Gill TP, Webb KL. Consumption of “extra” foods by Australian children: types, quantities and contribution to energy and nutrient intakes. Eur J Clin Nutr. 2008;62(3):356---364. 39. Rangan AM, Schindeler S, Hector DJ, Gill TP, Webb KL. Consumption of “extra” foods by Australian adults: types, quantities and contribution to energy and nutrient intakes. Eur J Clin Nutr. 2009;63(7):865---871. 40. Malik VS, Hu FB. Sweeteners and risk of obesity and type 2 diabetes: the role of sugar-sweetened beverages. Curr Diab Rep. 2012;12(2):195---203. 41. Nelson MC, Story M, Larson NI, Neumark-Sztainer D, Lytle LA. Emerging adulthood and college-aged youth: an overlooked age for weight-related behavior change. Obesity (Silver Spring). 2008;16(10):2205---2211. 42. Smith S. Conceptualising and evaluating experiences with brands on Facebook. Int J Market Res. 2013;55(3): 357---374. 43. Hamill S, Turk T, Murukutla N, Ghamrawy M, Mullin S. I “like” MPOWER: using Facebook, online ads and new media to mobilise tobacco control communities in low-income and middle-income countries. Tob Control. 2013;Epub ahead of print. 44. Hello Sunday Morning. Facebook. 2009. Available at: https://http://www.facebook.com/hellosundaymorning. Accessed February 10, 2014.

American Journal of Public Health | December 2014, Vol 104, No. 12