In March, a group of researchers inside Facebook compiled a report for one of the company’s most powerful executives, chief product officer Chris Cox.
The paper included a series of charts and data highlighting a troubling trend that seemed to be accelerating: Facebook was losing popularity with teens and young adults.
One colourful graphic showed that ‘time spent’ for US teenagers on Facebook was down 16% year-over-year, and that young adults in the US were also spending 5% less time on the social network.
The number of new teen signups was declining, and perhaps most concerning was a series of slides showing that young people were taking much longer to join Facebook than they had in the past.
Most people born before 2000 had created a Facebook account by age 19 or 20, the research showed. Now Facebook wasn’t expecting people born later to join the social network until they were much older, perhaps 24 or 25, if ever.
The report is among hundreds of internal documents collected by former Facebook employee-turned-whistle-blower Frances Haugen, who went public in early October with accusations that Facebook has been prioritizing profits over user safety and security.
The documents were disclosed to the U.S. Securities and Exchange Commission and provided to Congress in redacted form by Haugen’s legal counsel. The redacted versions were obtained by a consortium of news organizations, including Bloomberg News.
Some of the documents were previously used in stories on growth, teenagers and other topics by The Wall Street Journal.
Facebook shares fell less than 1% as the market opened in New York on Monday.
Taken together, the documents paint a stark picture of the years-long decline in growth metrics for key user groups, like teens and young adults in the U.S., on the flagship Facebook app. The internal reports, some of which haven’t previously been reported, show:
- Young people are spending less time on the service.
- Fewer teens are signing up.
- Many new teen accounts are duplicates, rather than unique new users.
- Users across age groups are creating fewer posts.
- Despite detailed research, employees don’t fully understand why these trends are happening or why product changes have failed to reverse them.
“What should we (particularly) optimize for among young adults?” one of the reports asked, referring to users between the ages of 18 and 29. “We don’t know enough to know.”
While Facebook has spent years studying its declining popularity among young people—an erosion that threatens the company’s advertising business, which has conquered an estimated 23.7% of the global digital ad market – Facebook executives have been markedly less forthcoming about those concerns in public.
The company’s tremendous business success has masked the persistent issues with young people. Facebook’s total audience has expanded consistently for years, and revenue has risen to nearly $30 billion per quarter. The company’s market value is close to $1 trillion.
Those successes mean the declines among teens and young adults on Facebook have been almost invisible to outsiders. Facebook doesn’t break down its user numbers by age group, and it has evaded questions from Wall Street analysts about the main app’s popularity with young people.
That discrepancy is the foundation of a formal whistleblower complaint filed by Haugen with the SEC.
Among Haugen’s arguments is that Facebook “has misrepresented core metrics to investors and advertisers” for years by showing overall growth but excluding details that show slowdowns in key demographics, according to the letter outlining her complaint.
Attracting teens, who make up an entire generation of consumers, has long been a challenge for Facebook. But user growth among a slightly older cohort—which Facebook refers to as ‘young adults’ – has been on the descent for almost a decade, too, the documents show.
The number of young adults on Facebook in the U.S. has declined 2% since 2019, and is expected to continue falling by an additional 4% over the next two years. Those young adults are also sharing less than they were a year ago, and sending fewer messages.
“There is a [young adult] sharing problem,” reads one internal report from early 2021. “They are choosing other apps to share day-to-day moments and life moments.”
Haugen also cited documents that show many Facebook and Instagram profiles are secondary accounts owned by a single person, an issue that she claims led Facebook to misrepresent its total audience size to advertisers.
A 2017 internal study found that more than 15% of all new Facebook accounts created by teenagers were second accounts—what the company calls SUMAs, or “Single User Multiple Account.”
In the US, 11% of all teen Facebook accounts were SUMAs as of January 2018, according to one report. That amounted to more than 1 million accounts.
All of these variables, from declining user engagement to duplicate accounts, suggest that Facebook hasn’t been transparent with investors and advertising clients about its core business, Haugen alleges. She says the documents she gave to the SEC and Congress prove it.
Our products are still widely used by teens, but we face tough competition from the likes of Snapchat and TikTok,” said Joe Osborne, a Facebook spokesperson.
“All social media companies want teens to use their services. We are no different. That’s why we’re continuing to build new products and features that are entertaining and help teens, their friends and family stay connected to each other.”
Osborne added that the company’s SEC filings include details about its challenges, including user engagement and estimates of duplicate and false accounts.
Facebook is “confident that our disclosures give investors the information they need to make informed decisions,” he said.
Facebook’s user growth and engagement are the company’s most important selling points for investors and advertisers.
The more users Facebook has, the more people it can target with advertisements. The more users interact with Facebook, the more ads those users are likely to see.
This makes Facebook’s monthly user base of almost 3 billion people—and the promise of its consistent future growth—a major driver of a 74% stock increase in the past two years.
Not all users are created equal, though. Facebook makes almost 13 times more money per user in the US and Canada than it does the Asia-Pacific region, for example.
Teenagers are also important. They don’t typically have established brand preferences in the way adults do, and young people often set cultural norms.
That’s why Haugen believes that Facebook has violated U.S. securities laws by failing to disclose to investors that key user cohorts – including teens and young adults – have been declining in total size and usage.
An internal chart included in Haugen’s filing with the SEC shows that Facebook’s teenage and young adult user bases in the US have shrunk since 2012, and more recent internal data show declines in time spent and expected user growth for both age groups.
That information could impact Facebook’s advertising appeal, Haugen said in her letter to regulators. “Facebook’s stock valuation is based almost entirely on predictions of future advertising growth,” the letter to the SEC reads.
Federal securities laws require that companies be truthful when making statements to shareholders, and Facebook has already been accused of violating those demands in recent years. In 2019, the social media giant paid $100 million to settle SEC allegations that it made misleading statements about the misuse of user account data by Cambridge Analytica, a consulting firm hired by former President Donald Trump’s campaign.
The regulator claimed that Facebook’s public disclosures described the potential misuse of customer data as a hypothetical when the company already knew such information had been improperly accessed.
The crucial test the SEC relies on in determining whether something should be disclosed is whether it’s material, meaning there is a substantial likelihood that a reasonable person would consider it important.
In reality, how materiality is applied is often murky, with no black or white guidelines.
For instance, companies have sometimes interpreted material matters as those that would have a specific impact on earnings, such as something that could lower profits by at least 5%.
But the SEC has said such assessments alone aren’t adequate for determining whether items are relevant to investors.
The recent whistle-blower allegations will likely put Facebook in the SEC’s cross-hairs again, said James Cox, a professor at Duke University School of Law who focuses on securities regulation.
The SEC will probably scrutinize the company’s public statements on user engagement and how they compare to internal communications, he said.
“If I were at the SEC, I’d map out the statements Facebook made over time and then line them up with the documents from the whistle-blower,” Cox said. “You’re going to find a lot of round pegs that don’t fit into square holes.”
Facebook acknowledged a teen retention issue on an earnings call in 2013, according to a review of transcripts by Bloomberg. In October of that year, then-Chief Financial Officer David Ebersman said the company had noticed a decline in usage among “younger teens.”
“Usage of Facebook among US teens overall was stable from Q2 to Q3,” he said. “But we did see a decrease in daily users, specifically among younger teens.”
That admission helped erase post-market stock gains after an otherwise solid quarter, and came less than a year after Facebook tried to buy rival Snapchat for a reported $3 billion.
Facebook instead focuses on data points each quarter that tend to show the company’s overall user base is growing. Facebook reports the total number of “monthly active people” and “daily active people” across all its apps, but it doesn’t report how many users it has by country, or how old those users are.
In other instances when analysts have asked about teens on quarterly earnings calls, Facebook has declined to share specifics. In July 2016, shortly after Facebook created an internal team focused specifically on teen-related products, CFO Dave Wehner was asked to share details on trends Facebook was seeing with younger users.
“On the teen front, younger users, we continue to be the best way to reach the largest global audience of teens and millennials,” he replied. “Teens remain very engaged on Facebook.
Clearly, how they’ve used our service has evolved over the years. And in addition to Facebook, they’re using Instagram, Messenger and WhatsApp. So, from a teen perspective, that’s some color there.”
In her letter to the SEC, Haugen also says Facebook is mischaracterizing its growth by failing to disclose that many accounts belonging to teenagers are actually duplicate accounts.
These accounts have been disclosed generally by Facebook, and represent a key data point for investors and advertisers because they can lead to inflated estimates about the number of people who are actually using the service.
The company’s quarterly earnings filings show that duplicate accounts represent about 11% of its global monthly active user base, but one internal Facebook analysis found that the company’s model undercounts the number of so-called SUMAs.
That metric is particularly important when it comes to counting teenagers. According to a 2017 document, more than 15% of new teen accounts are being created by existing users.
More than half of teens who create additional profiles aren’t active on multiple accounts. That’s unlike so-called Finsta accounts on Instagram, often set up by younger people to communicate privately with a smaller circle of friends.
Facebook’s SUMAs are more like account resets rather than separate accounts, according to the 2017 document.
Duplicate accounts on Facebook are leading to “extensive fraud” against advertisers, Haugen alleges. Advertisers pay Facebook to target their campaigns at specific audiences at a certain frequency.
Duplicate accounts raise the risk that advertising aimed at different accounts are actually directed at the same person, and that advertisers are overcharged, Haugen says.
A 2018 analysis on duplicate accounts from the internal documents estimated that a majority of so-called reach and frequency advertising campaigns using broad audience targeting would see an audience reduction of 5% to 8% when SUMAs are accounted for.
The report characterizes this effect as “limited.” A company spokesperson said that Facebook discloses that some ads reach duplicate accounts in the company’s “Help Center.”
The issue of duplicate accounts and the true size of Facebook’s user base is already the subject of a lawsuit against the company.
In a complaint filed this year in federal court in San Francisco, advertisers claim that Facebook inflated the number of people they could reach on the platform, partly because of duplicate accounts, leading to overcharges.
As teenage usage in the U.S. was declining in early 2016, Facebook spun up a “Teens Team” internally to build products focused on young people.
The group was small to start, about a half-dozen people, but it eventually grew to around 100 by late 2017. It built a number of features targeted at teens, including a video app for people under 21 called Lifestage and led video efforts to entice young creators in international regions, like the Middle East and Latin America.
The team, eventually called the “Youth Team,” also ran Messenger Kids, a chat app designed for elementary-school children.
Members of the team even contributed to a report for Chief Executive Officer Mark Zuckerberg in July 2016 suggesting that Facebook could buy Musical.ly, the viral-video sharing app that became TikTok, according to two former employees.
The documents illustrate how Facebook may have been grasping for ways to turn around the app’s fortunes with young people, conducting extensive research into their habits and perceptions of the company and brainstorming ways to boost engagement.
One research report reviewed by Bloomberg shows Facebook tested different messaging features, like emojis and face filters, with children as young as 6.
Michael Sayman, who worked on the “Teens Team” after joining Facebook when he was 18, would present internally on teen-related topics for other employees interested in a young person’s perspective.
“The whole company was just trying to understand a generation that they weren’t a part of,” said Sayman, who has since written a book that includes his experiences at Facebook.
Despite those efforts, Facebook is still struggling to attract teens and young people to its flagship app. Internal data from 2021 shows that for both age groups, time spent on the app, content production, messages sent and retention, or new users maintaining their account 90 days after its creation, are all declining.
The table, which was included in the “Young Adults and Teens” report prepared for Cox last March, identified all four of these categories as either “concerning” or “problematic.”
Part of the issue, the company found in its research, is that Facebook doesn’t offer young people a clear reason to use the network. In a recent study, Facebook asked young adults which apps they use for 12 different activities across video, commerce, news and entertainment.
Facebook was the top choice in just one category: “Get local information or connect with people in your area.”
“Overall there’s not strong value prop for FB among [young adults],” that same report reads. Another slide shows that while young adults want to be “uplifted and motivated” by content they see, they perceive “content on FB as negative, fake and misleading.”
For almost a decade, Facebook has countered its core social network’s relative weakness with teens by leaning on Instagram, the photo and video app it purchased for $1 billion in 2012.
When it comes to attracting teens, Instagram is the company’s main priority, documents show, and the goal is for those teens to graduate, or “age up,” to the main Facebook app as they become adults.
But in recent years, that process of “aging up” has started to slow. And while Instagram’s total user base is “saturated” in the company’s top five markets, meaning the app has reached the vast majority of potential users, time spent per user was down in all of them compared with the previous year, according to a report from March 2021.
That includes a 5% decline in the U.S. and an 8% decline in Japan, its two largest regions by users.
Content creation was also down among Instagram teens in all five markets, and messaging dropped in the three markets the company had data for: The US, Japan and Australia. The chart labels time spent and content creation as “problematic” for all five countries.
Facebook is also contending with increasingly robust competition from several other teen-centric apps.
The March 2021 internal report found that users are spending more than twice as much time on ByteDance Ltd.’s TikTok than they are Facebook. It also estimated that teens spend “2-3X more time” on TikTok than they do on Instagram, despite recently adding a TikTok clone inside Instagram, called Reels.
Other data from May 2021 shows that, among 16- to 24-year-olds, Facebook had slower user growth than Reddit Inc., TikTok, Twitter Inc., Discord Inc. and Snap Inc.’s Snapchat.
Instagram remains Facebook’s best bet with teens, and is still the focus of its strategy to lure in the next wave of internet users. One internal study found that teens can “shape the household’s perception of Instagram” based on how they use and talk about the app, meaning teen users can be a valuable vehicle for getting their siblings and parents to join up.
Still, that family influence could have paradoxical affects. “Teens strongly influenced preteens’ understanding of what and how frequently to share on Instagram,” according to one internal report, which added that teens could even discourage siblings from sharing because of “permanence” or the idea that “being spontaneous/authentic doesn’t belong on IG.”
“If this holds at scale,” the report continues, “teens could be creating sharing barriers for upcoming generations.”