The Instagram Reels Gold Rush

By Jacob Sweet

When Christian Koch and Aren Andersen began making pop songs together, in early 2020, most everyone in Nashville gave them the same advice: coming up with quality songs was great, but if they wanted traction for their band, they needed to post on TikTok and Instagram. The content was less important than the frequency; the gold standard was four posts a day. Though the advice didn’t exactly testify to music’s transcendent power, it made some sense to Koch.

Koch, who is twenty-five, had been promoting his solo music on Instagram for years. By October of 2021, he had more than twenty thousand followers, an audience that was large enough for Instagram to select him for the Reels Play Bonus, a new program that paid creators for the views they got on short-form videos known as Reels. Koch ignored the offer at first, but changed his mind and looked at what others were posting on Reels. The answer, in large part, was face-filter videos. Countless clips of people gazing at the camera as a question floated above their heads: “What Disney princess are you?” “How many children will you have?” “What astrological sign is your true love?” A few seconds later, the filter reveals a randomly generated answer (“Snow White!” “Twelve!” “Sagittarius!”) and those in front of the camera react as if they had won the lottery or sprained an ankle.

After spending a few minutes one night making face-filter videos, Koch woke up the next morning to find money waiting for him in his Instagram account. “I was, like, Oh, my God, I made ten bucks for doing that?” he recalled. “That’s awesome.” He began posting six to eight face-filter videos a day, and, weeks later, after receiving advice from mentors in the music industry to post more, upped his daily production to between sixteen and thirty. The Instagram algorithm rewarded this extraordinary proliferation of content. A filter that covered half his face with his random celebrity twin (Margot Robbie) got nearly forty-eight million views. During his second month enrolled in the new bonus program, he hit his maximum monthly allowance—a thousand dollars—and watched his follower count climb by thousands. When he and Andersen began churning out face-filter videos for their new band, Instagram granted them a maximum monthly bonus of thirty-five thousand dollars. For good measure, Andersen began posting face-filter videos on his personal account, too.

The duo had exploited a trend that still baffles them almost a year later. They don’t know who watches their videos and why they choose to do so. Nor do they understand why Instagram pushes their low-effort content to millions of people. But as long as fifteen minutes a day allows them to forgo full-time jobs and focus on music, they will continue to pump out face-filter videos. “I do it every day to make sure I can pay my rent, dude,” Koch said.

Koch and Andersen were not the only ones who jumped on the trend. Within a few months of the Reels Play Bonus announcement, dozens of others—from popular YouTube and TikTok creators to a recently divorced mother with no previous online following—seized the opportunity. By mass-producing face-filter videos, aspiring creators who had previously struggled to find an audience on Instagram saw their accounts balloon to hundreds of thousands of followers and their videos reach hundreds of millions per month.

Drew Beilfuss had posted skateboarding videos for six years before he gained consistent online viewership. But as his TikTok account rose toward a million followers in 2020, Beilfuss faced constant pressure from the app, which repeatedly took down his videos for containing what was termed “Dangerous Acts.” He looked to Snapchat, which, in an attempt to compete in the short-video realm, was paying out a million dollars a day to creators who posted viral content on its new TikTok clone, Snapchat Spotlight. In the first ten months of the program, launched in November, 2020, several creators had capitalized on the relatively low competition and earned millions of dollars for creating short, catchy videos. In June of 2021, Beilfuss belatedly realized that he could compete and began posting thirty videos a day. After a few dry weeks, he received multiple five-figure bonuses. In about a month, he earned more than enough to buy a Tesla.

Despite the huge profit Beifluss made off this new feature, he couldn’t help feeling disappointed. Had he started making videos when the program began, he could have made much more. A few months later, the Reels Play Bonus was his next chance—and face filters were the key to his success. After waking up and showering, he would hop on Instagram, click through trending Reels and sounds, and post ten face-filter videos back to back, exaggerating his reactions so they translated to video. Within a few weeks, Instagram was blasting out his face to millions of people. “I had some friends text me and they were, like, ‘What are you doing?,’ ” he told me. Evan Schaben, a red-headed friend of Beilfuss’s, pursued the same strategy on an account with zero preëxisting followers. After a couple of tepid months, some of his videos started hitting millions of views. He was leaving his orthodontist’s office when Instagram told him that he was eligible for a thirty-five-thousand-dollar monthly bonus; his consistent monthly view counts over a hundred million entitled him to about half the total over the next few months. His first purchase was Invisalign. If Instagram was going to make people look at his face, he didn’t want them making fun of his teeth.

Beilfuss and Schaben found that whether or not people liked the videos didn’t affect their popularity. As they posted over-the-top reactions to filters that guessed their iPhone model or told them how attractive they were on a scale of one to ten, they received frequent comments from people begging that Instagram stop showing their faces on their Discover pages. Many weighed in to tell them that they were making “the worst content ever” and responded with pleas to “#StopThisTrend.” The negative responses didn’t bother Beilfuss much. “I’m not gonna lie. That comment helps me.” For algorithmic purposes, engagement was engagement, even if it was someone engaging to say, “I’m ashamed to be alive.”

As a rising cohort of low-effort Instagram creators pumped out content, they handled these reactions differently. Some tried to make videos that people would hate less. Katie Feeney, a sophomore at Penn State whose several million TikTok and YouTube followers watch her sports and life-style content, tried to react genuinely to filters—avoiding the theatrical looks of disgust or stunned amazement that rile people up in the comments. She also mostly eschewed the popular but widely reviled trend of “3x videos,” in which creators encourage viewers to replay their videos at a faster rate of speed, promising a surprise that usually doesn’t materialize. Other creators hinted that they were in on the joke. Kelly Grace Richardson, a Georgetown undergraduate and former child actor, received twenty million views on a video that showed what she would look like if she dyed her hair blond. It turned out that she looked pretty similar, because her hair is already blond.

But some, recognizing that the algorithm’s preferences were out of their control, fully embraced content apathy. One creator, Brandon Phillips, who goes by Sk8bord B on social media, played an interactive filter game in which he had to identify the font color of as many words as possible. After correctly determining that the word “red” was written in the color blue, he looked at the word “orange,” shown in an unmistakable blue font, and picked the incorrect color. When the game told him he was wrong, he dropped his jaw in disbelief. It was all part of his plan; he had seen others get the first prompt wrong to invoke viewer rage, but he predicted waiting until the second would increase average watch time and draw more attention. His hunch was correct; the video got seventy-three million views. So inspired was Phillips that he did the same thing, with the same reaction, several more times. One video eclipsed eight million views, another five million.

Like several other face-filter creators, Phillips had experience exploiting viral trends by the time he got to Instagram. On TikTok, he was one of the most prolific creators of “Hide and Seek videos,” in which he would tell viewers to close their eyes, shrink himself with digital effects, and then disguise himself in a Where’s-Waldo-like landscape. The longer people lingered on his video—waiting for him to explain the concept and taking time to find him—the more TikTok promoted it. But he took pride in producing all types of videos: comedy skits, rap verses, messages written in code. He liked producing varied content, not fitting into any one niche. The app didn’t pay him much money, though, and he felt further discouraged from growing his account when TikTok demonetized and took down his videos for seemingly no reason.

When he heard about the Reels Play Bonus, in late 2021, he started reposting his most popular TikTok videos onto Instagram. After some didn’t resonate in quite the same way, he started studying Reels the way he’d once analyzed TikTok. What made Instagram promote a video? Was it the filter? The audio? The reaction of the person behind the camera? He kept track of popular filters and audio clips and tried them out himself when he had a few minutes to spare. He knew he was onto something when a video of him reacting to a teeth-whitening filter got sixty million views. Face-filter creators who spend hours editing videos for other platforms have also acquiesced to Instagram’s quantity-over-quality reality. Feeney, the Penn State student, at first tried posting the same kind of quality content that helped her grow large audiences on TikTok and YouTube, but found that face-filter videos performed much better. “The growth was just insane,” she told me. “It was nothing I had ever seen before.”

Every face-filter creator I spoke to posts their videos not on their main feed, which is sent out to all of their followers, but directly to the algorithmic Discover page. That way, the videos only go to people who have limited control over what Instagram shows them. For those who have seen success making content they’re more proud of, the logic behind making face filters is simple. Not only do the videos make money and improve the odds of getting lucrative brand deals but they can also direct people toward their other accounts, where they post better videos. But they also have come to understand how important it is to follow trends and game each site’s algorithm. Feeney, who gained a large chunk of her TikTok followers by embracing the ever-popular “unboxing” genre, was also one of the early adopters of Snapchat Spotlight—she told me that she made $1.6 million from the app. When YouTube Shorts launched domestically, in March, 2021, her rapid-fire posting on the company’s new and heavily promoted feature rocketed her to nearly three million followers in less than two years.

But for people who owe their first biggest online success to Instagram’s devout appreciation of low-quality content, this reality can be somewhat demoralizing. Monica Beck, a twenty-five-year-old model and content creator from Los Angeles, spent the pandemic streaming on Twitch for hours a day, hoping to create a community of people who cared about mental health and wellness. A good day would mean a couple hundred concurrent viewers. On Instagram, her first face-filter video got millions of views, so she decided to do the same thing over and over again. A few months later, she had more than four hundred thousand followers, about ninety per cent of whom began following her after her face-filter deluge. For a while, it felt disempowering to achieve so much recognition for something that had so little meaning and impact; it wasn’t the way she’d imagined growing an audience. “I think for a while, I was kind of judging it and negative about it,” she told me. “Until I kind of just realized, O.K., this is the nature of social media right now. You either ride the wave or not, and me not riding the wave led to me not getting any benefit out of it.” People could call her videos “cringe” all they wanted, but once she had the audience, she could finally start forming the community she’d wanted to grow in the first place.

On a Friday evening in Washington State, Brandon Phillips was hard at work. He scrolled through his saved videos on his Instagram, looking for something to post. He’d had a long day setting up acoustic panels in his new recording studio as he prepared to open it to the public, but now he had a few minutes to himself. He clicked a filter he hadn’t done yet, “Somehow this filter knows your zodiac sign,” and checked out others who have used it. They’d received three hundred thousand views, eighty-eight thousand, forty-three thousand, a hundred and fifty thousand. He decided to go for it, too. He dimmed his ring light, made sure he was right at the center of the frame, and set his camera’s auto-timer for three seconds. Then he let it rip.

He was hoping for Aquarius, but he got Sagittarius. On his second try, Sagittarius again. “I won’t waste too much time,” he told me. “I’m not gonna try five hundred times to get Aquarius.” If need be, he could just change the text below his face to “This filter knows your soulmate’s zodiac sign,” he said. “It usually still ends up going viral.” Luckily, the filter guessed correctly on the fourth try. Phillips’s eyes bulged as his mouth widened into an “O.” He covered his mouth like he’d walked unsuspectingly into a surprise birthday party. It was textbook execution. When he finished, he tightly cropped the video so viewers might accidentally start watching it a second time, then added a caption: “how did it know??” He made sure the video was hidden from his main feed, and, most important, made sure it counts toward his Reels Play Bonus. “It could go viral,” he said, “or it could go absolutely nowhere.”

He noted that over the last few weeks, his Reels hadn’t been doing as well as usual. Creators of all kinds were saying the same thing; they didn’t know what was up with the algorithm. Phillips had been getting recommendations to watch month-old videos for reasons he couldn’t explain. It was not the most promising start for someone who had just quit his full-time job, but he wasn’t going to stop now. “Some people get discouraged,” he told me. “But, you know, what can you do? I’ve seen slow starts that turned into good endings.” Weeks later, the video had more than two million views. The algorithm was on his side, prompting the work he cared about the least. ♦