ASU professor's article asks: Can social media users tell when they're being advertised to?
Download mp3 (1.82 MB)
Social media use has grown exponentially in the last decade, and the explosive launch of e-commerce has given rise to a new era of advertising: influencer marketing.
Which has led researchers to ask: Can users tell when they’re being advertised to? And how important is it for them to know when influencers are being paid?
David Franklyn is a law professor at Arizona State University and executive director of the McCarthy Institute. He co-authored an article that looks at influencer content on Instagram and TikTok.
“We found that some people can, some people can’t,” said Franklyn. “It depends on the type of content, but in many instances, over half of the people who are watching various videos got it wrong. You know, they thought that a paid one was unpaid, or that an unpaid one was paid.”
The Federal Trade Commission requires online creators and influencers to clearly disclose when they’re paid to post.
“They’re not doing it, I would say 60% of the time or more,” Franklyn said. “Many of them are doing it. Many of them are not.”
Labels are often vague, hard to find or nonexistent.
“The labels that are used like ‘sponsored’ or ‘hashtag ad,’ they're not curing this confusion,” said Franklyn.
Influencer marketing started with social media users over a decade ago posting product reviews, some of which would gain traction and attract millions of views.
Once brands discovered the value of marketing through influencers, review content became a growing niche and a large contributor to carving out the influencer or social media content creator career path.
It started to become more common for brands to send an influencer free products, provide free services, and in some cases pay thousands of dollars in exchange for a sponsored post.
“I think the shift in the phenomenon in the last five has been that it’s becoming more top-down and less organic,” said Franklyn. “Big business has discovered this and channeled it into systems that make it optimally economically beneficial.”
Which sometimes means curating the messaging their product appears alongside.
Brands that work with influencers have been known to include terms in their contracts that can range from keywords or talking points to the content being fully scripted by the company.
“The brand is basically finding an effective influencer who might have made herself famous partially on her own, by developing a following,” Franklyn said. “But now she’s become, at least in part, a paid actor for the brand. The content is written either by the brand owner’s marketing department or a company they hire.”
Franklyn said the way we browse social media — increasingly by scrolling through short-form, fast-paced content — means many users have very little context to judge whether what they’re seeing is sponsored or not.
“How can you consider the source when you don’t know what the source is?” said Franklyn. “We need more pressure on people to disclose the source of the information that they’re providing.”
One way, he said, is to know what brand owners are responsible for disclosing and putting pressure on them to be transparent about how they interact with influencers.
“If we’re gonna try to restore some sense of source identification, fidelity to communications,” Franklyn said, “We need to think about where to put the pressure in the economic and information ecosystem. That’s a point we try to make in the paper.”
Moving forward, Franklyn said looking at the issue through an intergenerational lens will also be key to designing solutions around what kind of regulation people want to see.
“Some of that correlated with age,” Franklyn said of concerns about companies’ duty to disclose. “Middle-aged people and older people wanted disclosure. A lot of younger people were so accustomed to this — swimming in the sea of, you know, the Internet, that they didn’t seem to care as much.”
But Franklyn argues that blurring the lines between paid and unpaid content disempowers internet users and could have broader implications on people’s ability to identify reliable information.
