social media social experiment

• Share full article

Advertisement

Supported by

LinkedIn Ran Social Experiments on 20 Million Users Over Five Years

A study that looked back at those tests found that relatively weak social connections were more helpful in finding jobs than stronger social ties.

By Natasha Singer

Natasha Singer, a business reporter at The New York Times, teaches a tech accountability journalism course at The Times’s summer program for high school students.

Listen to This Article

LinkedIn ran experiments on more than 20 million users over five years that, while intended to improve how the platform worked for members, could have affected some people’s livelihoods, according to a new study.

In experiments conducted around the world from 2015 to 2019, Linkedin randomly varied the proportion of weak and strong contacts suggested by its “People You May Know” algorithm — the company’s automated system for recommending new connections to its users. Researchers at LinkedIn, M.I.T., Stanford and Harvard Business School later analyzed aggregate data from the tests in a study published this month in the journal Science.

LinkedIn’s algorithmic experiments may come as a surprise to millions of people because the company did not inform users that the tests were underway.

Tech giants like LinkedIn, the world’s largest professional network, routinely run large-scale experiments in which they try out different versions of app features, web designs and algorithms on different people. The longstanding practice, called A/B testing, is intended to improve consumers’ experiences and keep them engaged, which helps the companies make money through premium membership fees or advertising. Users often have no idea that companies are running the tests on them. (The New York Times uses such tests to assess the wording of headlines and to make decisions about the products and features the company releases.)

But the changes made by LinkedIn are indicative of how such tweaks to widely used algorithms can become social engineering experiments with potentially life-altering consequences for many people. Experts who study the societal impacts of computing said conducting long, large-scale experiments on people that could affect their job prospects, in ways that are invisible to them, raised questions about industry transparency and research oversight.

“The findings suggest that some users had better access to job opportunities or a meaningful difference in access to job opportunities,” said Michael Zimmer , an associate professor of computer science and the director of the Center for Data, Ethics and Society at Marquette University. “These are the kind of long-term consequences that need to be contemplated when we think of the ethics of engaging in this kind of big data research.”

We are having trouble retrieving the article content.

Please enable JavaScript in your browser settings.

Thank you for your patience while we verify access. If you are in Reader mode please exit and  log into  your Times account, or  subscribe  for all of The Times.

Thank you for your patience while we verify access.

Already a subscriber?  Log in .

Want all of The Times?  Subscribe .

The Dangerous Experiment on Teen Girls

The preponderance of the evidence suggests that social media is causing real damage to adolescents.

S ocial media gets blamed for many of America’s ills, including the polarization of our politics and the erosion of truth itself. But proving that harms have occurred to all of society is hard. Far easier to show is the damage to a specific class of people: adolescent girls, whose rates of depression, anxiety, and self-injury surged in the early 2010s, as social-media platforms proliferated and expanded . Much more than for boys, adolescence typically heightens girls’ self-consciousness about their changing body and amplifies insecurities about where they fit in their social network. Social media—particularly Instagram, which displaces other forms of interaction among teens, puts the size of their friend group on public display, and subjects their physical appearance to the hard metrics of likes and comment counts—takes the worst parts of middle school and glossy women’s magazines and intensifies them.

One major question, though, is how much proof parents, regulators, and legislators need before intervening to protect vulnerable young people. If Americans do nothing until researchers can show beyond a reasonable doubt that Instagram and its owner, Facebook (which now calls itself Meta), are hurting teen girls, these platforms might never be held accountable and the harm could continue indefinitely. The preponderance of the evidence now available is disturbing enough to warrant action.

Facebook has dominated the social-media world for nearly a decade and a half. Its flagship product supplanted earlier platforms and quickly became ubiquitous in schools and American life more broadly. When it bought its emerging rival Instagram in 2012, Facebook didn’t take a healthy platform and turn it toxic. Mark Zuckerberg’s company actually made few major changes in its first years of owning the photo-sharing app, whose users have always skewed younger and more female . The toxicity comes from the very nature of a platform that girls use to post photographs of themselves and await the public judgments of others.

From the September 2017 issue: Have smartphones destroyed a generation?

The available evidence suggests that Facebook’s products have probably harmed millions of girls. If public officials want to make that case, it could go like this:

1. Harm to teens is occurring on a massive scale.

For several years, Jean Twenge, the author of iGen , and I have been collecting the academic research on the relationship between teen mental health and social media . Something terrible has happened to Gen Z, the generation born after 1996. Rates of teen depression and anxiety have gone up and down over time, but it is rare to find an “elbow” in these data sets––a substantial and sustained change occurring within just two or three years. Yet when we look at what happened to American teens in the early 2010s, we see many such turning points, usually sharper for girls. The data for adolescent depression are noteworthy:

Some have argued that these increases reflect nothing more than Gen Z’s increased willingness to disclose their mental-health problems. But researchers have found corresponding increases in measurable behaviors such as suicide (for both sexes), and emergency-department admissions for self-harm (for girls only). From 2010 to 2014, rates of hospital admission for self-harm did not increase at all for women in their early 20s, or for boys or young men, but they doubled for girls ages 10 to 14 .

The Facebook Papers: ‘History will not judge us kindly’

Similar increases occurred at the same time for girls in Canada for mood disorders and for self-harm . Girls in the U.K. also experienced very large increases in anxiety, depression, and self-harm (with much smaller increases for boys).

2. The timing points to social media.

National surveys of American high-school students show that only about 63 percent reported using a “social networking site” on a daily basis back in 2010. But as smartphone ownership increased, access became easier and visits became more frequent. By 2014, 80 percent of high-school students said they used a social-media platform on a daily basis, and 24 percent said that they were online “almost constantly.” Of course, teens had long been texting each other, but from 2010 to 2014, high-school students moved much more of their lives onto social-media platforms. Notably, girls became much heavier users of the new visually oriented platforms , primarily Instagram (which by 2013 had more than 100 million users ), followed by Snapchat, Pinterest, and Tumblr.

Boys are glued to their screens as well, but they aren’t using social media as much; they spend far more time playing video games. When a boy steps away from the console, he does not spend the next few hours worrying about what other players are saying about him. Instagram, in contrast, can loom in a girl’s mind even when the app is not open, driving hours of obsessive thought, worry, and shame.

3. The victims point to Instagram.

The evidence is not just circumstantial; we also have eyewitness testimony. In 2017, British researchers asked 1,500 teens to rate how each of the major social-media platforms affected them on certain well-being measures, including anxiety, loneliness, body image, and sleep. Instagram scored as the most harmful, followed by Snapchat and then Facebook. Facebook’s own research , leaked by the whistleblower Frances Haugen, has a similar finding: “Teens blame Instagram for increases in the rate of anxiety and depression … This reaction was unprompted and consistent across all groups.” The researchers also noted that “social comparison is worse” on Instagram than on rival apps. Snapchat’s filters “keep the focus on the face,” whereas Instagram “focuses heavily on the body and lifestyle.” A recent experiment confirmed these observations: Young women were randomly assigned to use Instagram, use Facebook, or play a simple video game for seven minutes. The researchers found that “those who used Instagram, but not Facebook, showed decreased body satisfaction, decreased positive affect, and increased negative affect.”

Read: Teens are bullied ‘constantly’ on Instagram

4. No other suspect is equally plausible.

Many things changed in the early 2010s. Some have suggested that the cause of worsening mental health could be the economic insecurity that followed the 2008 global financial crisis. But why this would hit younger teen girls the hardest is unclear. Besides, the American economy improved steadily in the years after 2011, while teen mental health deteriorated steadily. Some have suggested that the 9/11 attacks, school shootings, or other news events turned young Americans into “ generation disaster .” But why, then, do similar trends exist among girls in Canada and the U.K.? Not all countries show obvious increases in mood disorders, perhaps because technological changes interact with cultural variables, but the societies most like ours (including Australia and New Zealand ) exhibit much the same patterns.

Correlation does not prove causation, but nobody has yet found an alternative explanation for the massive, sudden, gendered, multinational deterioration of teen mental health during the period in question.

T o be sure, there is evidence on the other side. Dozens of studies and several meta-analyses (studies of groups of studies) have examined the relationship between greater digital-media use and worse teen mental health, and most have found just small correlations, or none at all. The most widely cited of these studies, published in 2019, analyzed 355,000 teens across three large data sets from the U.S. and U.K. The authors found only a tiny correlation—no larger than the correlation of bad mental health with self-reports of “eating potatoes.” Facebook cites this research in its defense.

But here’s the problem with these studies: Most lump all screen-based activities together (including those that are harmless, such as watching movies or texting with friends), and most lump boys and girls together. Such studies cannot be used to evaluate the more specific hypothesis that Instagram is harmful to girls. It’s like trying to prove that Saturn has rings when all you have is a dozen blurry photos of the entire night sky.

Derek Thompson: Social media is attention alcohol

But as the resolution of the pictures increases, the rings appear. The subset of studies that allow researchers to isolate social media , and Instagram in particular, show a much stronger relationship with poor mental health. The same goes for those that zoom in on girls rather than all teens. Girls who use social media heavily are about two or three times more likely to say that they are depressed than girls who use it lightly or not at all. (For boys, the same is true, but the relationship is smaller.) Most of the experiments that randomly assign people to reduce or give up social media for a week or more show a mental - health benefit , indicating that social media is a cause, not just a correlate.

F acebook would have you believe that merely cutting back the time that teens spend on social media will solve any problems it creates. In a 2019 internal essay , Andrew Bosworth, a longtime company executive, wrote:

While Facebook may not be nicotine I think it is probably like sugar. Sugar is delicious and for most of us there is a special place for it in our lives. But like all things it benefits from moderation.

Bosworth was proposing what medical researchers call a “dose-response relationship.” Sugar, salt, alcohol, and many other substances that are dangerous in large doses are harmless in small ones. This framing also implies that any health problems caused by social media result from the user’s lack of self-control. That’s exactly what Bosworth concluded: “Each of us must take responsibility for ourselves.” The dose-response frame also points to cheap solutions that pose no threat to its business model. The company can simply offer more tools to help Instagram and Facebook users limit their consumption.

But social-media platforms are not like sugar. They don’t just affect the individuals who overindulge. Rather, when teens went from texting their close friends on flip phones in 2010 to posting carefully curated photographs and awaiting comments and likes by 2014, the change rewired everyone’s social life.

Improvements in technology generally help friends connect, but the move onto social-media platforms also made it easier—indeed, almost obligatory––for users to perform for one another.

Public performance is risky. Private conversation is far more playful. A bad joke or poorly chosen word among friends elicits groans, or perhaps a rebuke and a chance to apologize. Getting repeated feedback in a low-stakes environment is one of the main ways that play builds social skills, physical skills, and the ability to properly judge risk . Play also strengthens friendships.

When girls started spending hours each day on Instagram, they lost many of the benefits of play. (Boys lost less, and may even have gained, when they took up multiplayer fantasy games, especially those that put them into teams.) The wrong photo can lead to school-wide or even national infamy, cyberbullying from strangers, and a permanent scarlet letter. Performative social media also puts girls into a trap: Those who choose not to play the game are cut off from their classmates. Instagram and, more recently, TikTok have become wired into the way teens interact, much as the telephone became essential to past generations.

Evelyn Douek: 1 billion TikTok users understand what Congress doesn’t

Facebook’s researchers understand the implications of this rewiring. In one slide from an internal presentation on Instagram’s mental-health effects, the presenter notes that “parents can’t understand and don’t know how to help.” The slide explains: “Today’s parents came of age in a time before smartphones and social media, but social media has fundamentally changed the landscape of adolescence.”

S ocial-media platforms were not initially designed for children, but children have nevertheless been the subject of a gigantic national experiment testing the effects of those platforms. Without a proper control group, we can’t be certain that the experiment has been a catastrophic failure, but it probably has been. Until someone comes up with a more plausible explanation for what has happened to Gen Z girls, the most prudent course of action for regulators, legislators, and parents is to take steps to mitigate the harm. Here are three:

First, Congress should pass legislation compelling Facebook, Instagram, and all other social-media platforms to allow academic researchers access to their data . One such bill is the Platform Transparency and Accountability Act, proposed by the Stanford University researcher Nate Persily .

Second, Congress should toughen the 1998 Children’s Online Privacy Protection Act. An early version of the legislation proposed 16 as the age at which children should legally be allowed to give away their data and their privacy. Unfortunately, e-commerce companies lobbied successfully to have the age of “internet adulthood” set instead at 13. Now, more than two decades later, today’s 13-year-olds are not doing well. Federal law is outdated and inadequate. The age should be raised. More power should be given to parents, less to companies.

Third, while Americans wait for lawmakers to act, parents can work with local schools to establish a norm: Delay entry to Instagram and other social platforms until high school.

Right now, families are trapped. I have heard many parents say that they don’t want their children on Instagram, but they allow them to lie about their age and open accounts because, well, that’s what everyone else has done. Dismantling such traps takes coordinated action, and the principals of local elementary and middle schools are well placed to initiate that coordination.

Haugen’s revelations have brought America to a decision point. If public officials do nothing, the current experiment will keep running—to Facebook’s benefit and teen girls’ detriment. The preponderance of the evidence is damning. Instead of waiting for certainty and letting Facebook off the hook again, we should hold it and other social-media companies accountable. They must change their platforms and their ways.

​When you buy a book using a link on this page, we receive a commission. Thank you for supporting The Atlantic.

About the Author

More Stories

End the Phone-Based Childhood Now

Get Phones Out of Schools Now

The consequences of social media’s giant experiment

Subscribe to the center for technology innovation newsletter, tom wheeler tom wheeler visiting fellow - governance studies , center for technology innovation.

January 13, 2021

Facebook and Twitter have banned Donald Trump from their platforms. Fleeing to the right-wing alternative Parler has been effectively shut down by Amazon’s decision to cease hosting it on Amazon Web Services while both Android and Apple have removed it from their app stores.

The actions of Facebook and Twitter are protected by Section 230 of the 1996 Communications Act. This is the same Section 230 behind which social media companies have sheltered to protect them from liability for the dissemination of the hate, lies and conspiracies that ultimately led to the assault on the U.S. Capitol on January 6.

These actions are better late than never. But the proverbial horse has left the barn. These editorial and business judgements do, however, demonstrate how companies have ample ability to act conscientiously to protect the responsible use of their platforms.

Subsection (2) of Section 230 provides that a platform shall not be liable for, “Any action voluntarily taken in good faith to restrict access to or availability of material that any provider or user considers to be…excessively violent, harassing, or otherwise objectionable…” In other words, editorial decisions by social media companies are protected, as long as they are undertaken in good faith.

It is Subsection (1) that has insulated social media companies from the responsibility of making such editorial judgements. These 26 words are the heart of the issue: “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” That single sentence creates the current conundrum. If you are insulated from the consequences of your actions and make a great deal of money by exploiting that insulation, then what is the incentive to act responsibly?

These 26 words are the heart of the issue: “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”

The companies have shut down at least one liar and blocked an alternative pathway for his rants. Facebook and Twitter (and Google’s YouTube subsidiary) are protected by subsection (2) when they make this editorial decision. Why didn’t they do it earlier? The actions of the companies to deal with content on the apps, promotion of the apps and hosting of the apps have been decisive and warranted. The question is “why not until now?”

The social media companies have put us in the middle of a huge and explosive lab experiment where we see the toxic combination of digital technology, unmoderated content, lies and hate. We now have the answer to what happens when these features and large profits are blended together in a connected world. The result not only has been unproductive for civil discourse, but also that it represents a danger to democratic systems and effective problem-solving.

Dealing with Donald Trump is a targeted problem that the companies just addressed decisively. The social media companies assert, however, that they have no way to meaningfully police the information flowing on their platform. It is hard to believe that the brilliant minds that produced the algorithms and artificial intelligence that powers those platforms are incapable of finding better outcomes from that which they have created. It is not technological incapacity that has kept them from exercising the responsibility we expect of all other media, it is the lack of will and desire for large-scale profits. The companies’ business model is built around holding a user’s attention so that they may display more paying messages. Delivering what the user wants to see, the more outrageous the better, holds that attention and rings the cash register.

Thus far our political leaders have expressed concern about the effects of Section 230. Too often that activity has been performance for their base rather than progress towards a solution. As Congress takes up serious consideration of Section 230, here are a few ideas:

Social media companies are media, not technology

Mark Zuckerberg testified to Congress, “I consider us to be a technology company because the primary thing we do is have engineers who write code and build product and services for other people.” That software code, however, makes editorial decisions about which information to choose to route to which people. That is a media decision. Social media companies make money by selling access to its users just like ABC, CNN, or The New York Times .

There are well established behavioral standards for media companies

The debate should be over whether and how those standards change because of user generated content. The absolute absence of liability afforded by Section 230 has kept that debate from occurring.

Technology must be a part of the solution

When the companies hire thousands of human reviewers it is more PR than protection. Asking humans to inspect the data constantly generated by algorithms is like watching a tsunami through a straw. The amazing power of computers created this situation, the amazing power of computers needs to be part of the solution.

It is time to quit acting in secret

When algorithms make decisions about which incoming content to select and to whom it is sent, the machines are making a protected editorial decision. Unlike the editorial decisions of traditional media whose editorial decisions are publicly announced in print or on screen and uniformly seen by everyone, the platforms’ determinations are secret: neither publicly announced nor uniformly available. The algorithmic editorial decision is only accidentally discoverable as to the source of the information and even that it is being distributed. Requiring the platforms to provide an open API (application programming interface) to their inflow and outflow, with appropriate privacy protections, would not interfere with editorial decision-making. It would, however, allow third parties to build their own algorithms so that, like other media, the results of the editorial process are seen by all.

Expecting social media companies to exercise responsibility over their practices is not a First Amendment issue. It is not government control or choice over the flow of information. It is rather the responsible exercise of free speech. Long ago it was determined that the lie that shouted “FIRE!” in a crowded theater was not free speech. We must now determine what is the equivalent of “FIRE!” in the crowded digital theater.

Apple, Amazon, Facebook, and Google are general, unrestricted donors to the Brookings Institution. The findings, interpretations and conclusions in this piece are solely those of the author and not influenced by any donation.

Related Books

Mark MacCarthy

November 7, 2023

Nicol Turner Lee

August 6, 2024

Darrell M. West, John R. Allen

July 28, 2020

Related Content

John Villasenor

October 27, 2022

June 8, 2020

Niam Yaraghi

April 9, 2019

Internet & Telecommunications

Governance Studies

Center for Technology Innovation

August 8, 2024

Sam Manning

July 3, 2024

Lily Morse, Mike H. M. Teodorescu, Yakov Bart

July 2, 2024

Wondering if third-party editing tools are ruining your Instagram reach? We were curious, too, so we put on our lab coats and went to work.

Experiment: How Many Instagram Stories Should You Post?

Which earns more reach: a single Instagram Story per day or many Stories per day? Is there even a “right” number of daily Stories to post?

Experiment: Where Should Your Instagram Hashtags Go?

Do Instagram hashtags belong in the caption or comments? Does hashtag location even matter? Time to put on our lab coats and start testing.

Experiment: Do Long-Form X Posts Give You More Reach?

What gives you better reach on X: a good old-fashioned tweet thread or one long-form post? Here’s what we found.

Experiment: Do Longer LinkedIn Comments Drive More Connection Requests?

Does being original on LinkedIn generate faster growth? Here’s a thrilling social experiment that answers the question.

Experiment: What Happens When AI Plans Your Entire Social Strategy?

Is AI really coming for your job? We decided to find out.

How Much Does YouTube Pay Per View? We Tested It!

Yes, YouTube pays per view—but that doesn’t mean that a million video views will make you rich. We’ll explain.

Experiment: How Many Hashtags To Use on Reels?

While it’s possible to add up to 30 tags in your Instagram Reels caption, more isn’t always more. So what’s the perfect number?

Become a better social marketer.

Get expert social media advice delivered straight to your inbox.

Experiment: Do Traditional SEO Tactics Work on Social Media?

Can we use what we know about optimizing content for Google to reverse-engineer social SEO success? To the laboratory!

Experiment: Does AI Write Better X (Twitter) Posts Than Humans?

Can human-written social posts generate more engagement and reach than AI-written ones? It’s humans vs. robots in the ultimate showdown.

Experiment: Do Instagram Carousels Still Outperform Reels?

Are carousels still the top content format on Instagram, or have Reels displaced them for good? There’s only one way to find out.

Experiment: Does Custom Instagram Alt Text Improve Reach?

Everybody’s talking about social SEO, but how do you actually do it? We tested a theory to see if Instagram alt text makes a difference.

October 15, 2020

Why Social Media Makes Us More Polarized and How to Fix It

Research shows it’s the influencers, not the networks themselves, that amplify differences between us

By Damon Centola

Getty Images

Every time I log onto Facebook, I brace myself. My newsfeed—like everyone else’s I know—is filled with friends, relatives and acquaintances arguing about COVID-19, masks and Trump. Facebook has become a battleground among partisan “echo chambers.” But what is it about social media that makes people so polarized?

To find out, my colleagues and I ran a social media experiment in which we divided Democrats and Republicans into “echo chambers,” or small groups whose members affiliate with just one political party. Next, we picked the most polarizing issues we could think of: immigration, gun control and unemployment. We asked each participant what they thought of those issues, then let people talk to each other and revise their opinions. After several rounds of discussion and revision, we evaluated each group’s viewpoint.

To our surprise, the echo chambers did not make people more polarized, but less . After interacting in social networks with likeminded peers, each echo chamber—Republican and Democrat—had adopted a more moderate opinion.  All groups independently moved toward opinions that were closer to the opinions on the “opposite” side of the political spectrum.

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing . By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

As a scientist who studies networks, I’m used to being surprised by the results of my experiments. Technology has allowed us to access more information and data about people’s social networks, debunking many of our assumptions about human behavior. But even my team at the Network Dynamics Group was surprised: Why did our social media experiment find the opposite of what happens all the time in the real world of social media?

The answer lies in something social media has amplified: “influencers.”

By now, most of us have a fairly specific understanding of what an “influencer” is. The word conjures up a young, wealthy person whose lifestyle is sponsored by brands like Instagram, TikTok or YouTube. But the word has a very specific network science meaning. In social media, networks tend to be centralized: a small number of people, or perhaps just one person, at the “center” of the network is connected to lots of other people in the “periphery.” The multitudes in the periphery of the social network have only a modest number of connections, while the few—the so-called “influencers”—at the center of the network are connected to nearly everyone. This puts these people into the powerful position of being able to exert a disproportionate level of “influence” over the group.

By contrast, the networks used in our study were “egalitarian”—the opposite of centralized. In an egalitarian network, everyone has an equal number of contacts, and therefore influence, throughout the network.

The key feature of an egalitarian network is that new ideas and opinions can emerge from anywhere in the community and spread to everyone. But in centralized networks—like many social media sites—ideas are filtered through, or sometimes even blocked, by a powerful social influencer. As I show in my upcoming book Change: The Power in the Periphery to Make Big Things Happen , centralized and egalitarian networks have very different effects on partisan bias and the acceptance of new ideas.

In a centralized echo chamber, if the influencer at the middle shows even a small amount of partisan bias, it can become amplified throughout the entire group. But in egalitarian networks, ideas spread based on their quality, and not the person touting them. There is a lot of wisdom in network peripheries, in regular people with good ideas. When the social network enables those people to talk with each other, new thinking that challenge a group’s biases can take hold and spread.

To see how egalitarian networks might affect other kinds of contentious issues, we conducted another experiment with smokers and nonsmokers discussing the risks of cigarette smoking. The effects were the same as with the partisan study. Both groups moved toward a more accurate understanding of smoking risks. Moreover, when participants were interviewed after the study, they reported having developed higher opinions of the other. Both smokers and nonsmokers had come to view the other group as more reasonable and trustworthy about the risks of smoking. But the exchange of ideas and eradication of bias only works when networks are egalitarian.

The problem of partisan bias is exacerbated on social media because online networks are often organized around a few key influencers. This feature of social media is one of the main reasons why misinformation and fake news has become so pervasive. In centralized networks, biased influencers have a disproportionate impact on their community—enabling small rumors and suppositions to become amplified into widespread misconceptions and false beliefs. 

Our country has been struggling with bias and polarization a long time. But the issue is about to get much more urgent. As the debate over COVID-19 vaccination heats up, biased viewpoints will undoubtedly become entrenched in communities with powerful influencers at their center. If we want to eradicate, or at least lessen the impacts of the coronavirus, we should rethink how our online communities operate. The solution to the problem of vaccine hesitancy is not to eliminate echo chambers. Rather, it is to be intentional about the social networks in those echo chambers. The more equity in people’s social networks, the less biased and more informed groups will become—even when those groups start off with highly partisan opinions.

The Ultimate Guide to Social Testing

Published: September 07, 2021

As marketers, we know the importance of making data-driven decisions. The more information we have about our audience, the more we’re able to make effective marketing moves. 

In addition, having the numbers to back up the implementation of a marketing strategy is almost as important as the strategy itself. One of the ways to get this data is through social media testing, where you figure out what campaigns resonate most with your audience and help you meet your marketing goals.

In this post, learn how you can run a social media test to help you meet your marketing goals and discover high-quality tools that will help you do so. 

Social Testing

Social testing, also called social media testing, is an experiment that shows how your content is performing among your target audiences. They test different forms of the same post simultaneously and are measured by a pre-set goal. The goal, for example, might be to raise engagement or a boost in lead generation.

For instance, you might run a social test to learn if video campaigns are worth investing in on Facebook, so you create a post that measures impressions of an ad with and without a video attached. After the campaign, the interactions with your post will tell you if a video is a worthwhile investment for your brand, which makes this process so important.

The Benefits of Social Testing

Social media testing is important because it provides data-driven insights about your social media marketing activities. It allows you to analyze how different variables, like photo and video, affect performance.

Ultimately, social media tests provide data about how audience behavior can influence the structure of your campaigns. You’ll get a picture of what is successful for your brand, and you can create campaigns that you know will work. Instead of researching countless industry benchmarks, you’ll have concrete data specific to your business that comes from testing results. 

There are multiple types of social media tests you can run for your business, and we’ll discuss them next.

Types of Social Tests

Let's say you want to know how copy affects an international audience on LinkedIn. Or, that you want evidence of a landing page performing better with a different image. Maybe you're trying to identify if changing the tone of Instagram captions will lead to more audience engagement.

All of these scenarios are prime for social tests. They point out a problem that can be answered with data. This data would give insight about audiences interactions with brands on social media.

After identifying the goal, it’s time to pick the type of test. Let's go through the different types and when you might use them.

A/B tests are likely the most common form of social testing. They look at a variable between two content types, measure the outlined goal, and provide results. So, consider running an A/B test if you want to test a single, small variable that may alter audience behavior.

For example, run an A/B test if you want to test out different CTA buttons on a Facebook ad or experiment with a post's copy with/without emojis. The image below is an example of what an A/B test can look like. 

Image Source

Split tests are often confused with A/B tests. In essence, they do the same thing: test two content types based on a goal. The difference is that a split test is more general than an A/B test; they're used to determine big changes, and the two variants are often entirely different.

Use split testing if you want to know which layout of your Facebook Business page performs better. You can also run split tests to determine A/B test factors. For example, you can run a split test to determine which ad video cut you will use for a Sponsored Tweet, then A/B test different sections of the winning cut. The image below is an example of what a split test could look like on a business website. 

Multivariable Test

Multivariable tests work differently than the two previously mentioned types, as they work with multiple variables instead of one or two. You can run a multivariable test to determine which of four ads works best. Alternatively, you can run one to assess the different makings of a post, like images vs. copy vs. captions.

Run a multivariate test if you want to see the results of more than two different elements. For example, compare the caption, image, and CTA of a LinkedIn post to see what audiences are responding to, or look at three versions of a GIF to learn how audiences react to them. The image below is an example of a what a multivariate test set up could look like.

When you've picked out your social test experiment, make sure you're imploring best practices so the results are helpful.

Social Testing Best Practices

Social media testing can be extremely helpful — if you're executing them correctly. Otherwise, your test could be inaccurate, immeasurable, and ultimately, a waste of time.

The good news is that social tests aren't hard to create or run. The bad news is that if you aren't prepared to run one, your results won't be usable. Make sure that when you design your test, you follow these best practices:

1. Have one specific goal.

When your social test has one identifiable goal, everything else falls into place: variables, unit of measurement and time frame. To illustrate this, let's say your goal is to improve international engagement with your next Facebook ad.

With that goal decided, you can create the variables, so you decide to run an A/B test to determine which copy earns the most impressions. You estimate that because your impression ads usually run for a month, half that time would give measurable results.

You also know the direction of the ad, so your focused goal makes it easier to know what to look for during analysis. Impressions, for example, would be the metric to look at for the Facebook ad mentioned earlier.

2. Know who your audiences are.

If you decide to conduct a social test, your data will reflect the behaviors of a specific target market and how they engage with your messaging. 

Social testing is a great way to learn about the social media behavior of an audience segment. If you had little to no information about how your millennial audience would react to a new Instagram Story Ad, running a test would give you a data-driven answer.

Without knowing your audience, your data wouldn't apply to a defined set of your target market. You’d have insight, but it would be unclear how results relate to your different audience segments . 

3. Take note of your current performance.

Before you run the test, know how your current campaign is running or note previous results. Then, at the end of your experiment, you can compare results and make informed decisions. The previous report will give the background information and context needed to analyze the social test findings.

Even if your marketing goals for your social test are different than previous campaigns, it's still a good idea to refer to them for context. For instance, you may be testing for conversions rather than retargeting, but having an idea of what audiences prefer helps you structure your campaign.

4. Monitor your test periodically.

Don't leave your test as soon as it begins — monitor it so you can adjust accordingly. A test that's running for a month, for example, should be checked regularly for performance benchmarks.

If you use social test software or tools on social media channels, the report starts aggregating when your test starts. So, when you check-in, look at how you're tracking for your #1 goal. Additionally, take note of what other metrics you see and how they're performing.

You might find, for instance, that your conversion rate is low. You can monitor conversions for the duration of the test or make a minor tweak to try to improve performance. For the subsequent check-ins, you'll have an additional metric to take into account.

When the test concludes, you'll have the knowledge from previous checks to round out your perception of the completed report. Along with your intended goal, identify supporting metrics to understand how they work together.

5. Make your test timely.

How long should you run your test? Long enough to get the answer to your hypothesis. That doesn't tell you much, though, so let's add to that.

Ideally, your test should run for at least seven days. A week is enough time for your social testing software to compile a basis of data. It won't be as concrete as a more extended test, but it's a starting point.

After seven days, look at your performance and decide if you've gathered enough data to answer your hypothesis. If not, run the test for a few more days. Then, based on the nature of your campaign, fix the time frame to fit your business and your audience — but make sure you give yourself enough room for an actionable report.

Picking a test duration period ensures you won't be wasting money and time. Instead of having a test run for too long or too little, figure out your time frame and budget during the planning process.

So, with these best practices in mind, you're almost ready to run a successful social test. Before you get going, though, let's talk about where — and how — to do that.

Where can I run a social test?

There's a couple of avenues to explore when choosing where to run a social network test. If you're running a test specifically for social media, the channel you're using might have testing tools in their business software.

Running social tests on the corresponding media platforms is helpful because you don't have to track different channels during the test. Additionally, you won't have to worry about misconstrued data a third party might provide.

You can also run a social test using a CRM, like HubSpot . CRMs are a great choice if you want to test functions outside of social media — landing pages, emails, or other marketing activities. First, though, let's talk about social media offerings.

Facebook Social Testing

Facebook offers tools to run A/B and multivariate tests for ads. You can access these tests via Ad Manager . Ad Manager tells you which ads (or ad tests) are running and their status. When you click on a certain test, details and metrics open.

Multivariate tests on Facebook are similar, except they can be found in the Experiments section of your Facebook Business account. Though the social tests are located in different places, the process for setting them up is largely the same.

Social testing on Facebook is intuitive, especially if you're used to running Facebook Ads. After clicking "Create Ad," you'll be taken through a series of prompts to create, set metrics, and pick an audience. When you fill in the details, you publish it, and wait for the results.

Start small if you're getting the hang of things. Its features are customizable, so the choice is yours in terms of time frame, audience, copy, testing options, and metrics. As your familiarity grows, tests can be scaled.

As an example, after running a social test on Facebook, bone broth brand Kettle & Fire found a 14 point increase in brand awareness . The marketing team wanted to raise online sales with a video campaign, and wanted to find out which video length was favored by customers.

In a little less than a month, a 1.5 lift in purchasing consideration and sales conversions from the short video led to the answer. A Facebook test saved the team ad spend and informed them about their audience's Facebook preferences.

Facebook delivers social testing results in a downloadable report. The software determines the winner based on the metric(s) you chose when creating the test. For information about running a Facebook social testing, check out this post about testing on the platform.

Twitter Social Testing

Social testing on Twitter leads to creating tweets audiences will interact with and enjoy. You'll identify how they use the platform and know how to cater to their needs. Twitter's testing tools are best fit for creative, targeting, and brand awareness campaigns.

Twitter lets you social test Ads in an A/B format. To do this, after publishing the ad, go back into Ads , duplicate it, make the change, and publish the new variable. Track performance from the Tweet Activity Dashboard and campaign dashboard. 

Note that there's no way to set an official test using Twitter, so you'll have to analyze the data yourself. For instance, if you want to see if your new brand voice is taking flight, you'll probably want to look at the amount of clicks and impressions your Ad earned.

LinkedIn Social Testing

If you're not used to LinkedIn Ads , there's several different types. They're sorted based on campaign goal, like Conversion, or type of ad, like Carousel, Text, or Sponsored. It's a great idea to test LinkedIn Ads for promoting — your brand, event, or job listing.

Dynamic Ads, which show up on the ride side of a user's main feed are highly personalized. Under Dynamic Ads, there's several versions: Follower, Job, Content, and Spotlight. These versions specify what the goal of your ad is — so if you want to let audiences know your company is hiring, you would run a Dynamic Job Ad.

If you wanted to make a similar Dynamic Job Ad on LinkedIn, you can run a test to see if your listing ad is compelling to job seekers. You'll have the option to test image layouts, such as the company photo, copy, and central images, if applicable.

LinkedIn's marketing team uses social testing for content and event promotion, as well as account-based marketing . They ran a test for a webinar Dynamic Ad to gauge if audiences preferred an image of the speaker or the company's logo. According to one of LinkedIn's senior marketing managers, Cassandra Clark, results included a 326% lift in click-through rate in the ad with the speaker.

To access ad results, check your Conversion Tracking dashboard, which will show you page and audience activity from your ads. While LinkedIn doesn't have a formal testing feature, they do have an option for you to duplicate and tweak a portion of an ad, like Twitter.

CRM Social Testing

If you're not running a test on a social media channel, using a CRM is another worthwhile option. With a CRM, you can set up tests for website content, like landing pages. 

In HubSpot, you can run social tests for web pages simply by accessing your dashboard and going to one of your web pages. Click the "Actions" hyperlink next to one of your pages, and you'll find the option to "Run a test."

HubSpot lets you run an A/B or multivariate (Adaptive) test. As you're creating your test, you'll have the option to look at testing tips, like figuring out what to test. Since HubSpot has a drag-and-drop editor, and tons of modules to choose from, you'll have plenty of options. In the past, I've run tests for CTA buttons, images, headline copy, body text, and landing page layout through social media.

It's commonplace for CRMs to have tools for running social testing. Some offer a niche version of social testing; for instance, MailChimp's are for emails. When you decide to use a CRM for testing, do some research to determine which one will offer you the tools you need to succeed.

But, if you don't know where to begin looking for software that offers social testing, here's a list of tools for your benefit.

Social Media Testing Tools

• OptinMonster

Price: Free plan, or $800 for Marketing Hub Professional

With HubSpot's CRM, you can run A/B and multivariate tests. The tools are part of HubSpot's Marketing Software and are included in the Marketing Hub Professional and Enterprise plans. You can use HubSpot to test landing pages that are promoted through social media.

The CRM's drag-and-drop editor makes it easy to configure variables for tests. In addition, you’ll have optimization features to make your page shareable and user-friendly, like social media badges. That way, when you promote the page on socials, your followers can share them with one click.

HubSpot's testing tool is accessible from the landing page dashboard via "Run a test" from the drop-down menu. From there, you can choose your test type and get started. If you want an easy-to-use platform that lets you create and analyze social tests, HubSpot is a great choice.

2. OptinMonster

Price: Free plan, or $19-49/mo.

OptinMonster is a CRM that also offers built-in testing software. You can run A/B or split tests to increase conversions over time with tools that let you experiment with content, headlines, campaign triggers, styles, and layouts.

Once you set up the test, you can leave the software to do the work. It'll show your test content to website visitors and collect data on conversion rates. Your report will determine the number of impressions, acquisitions, and page visits.

You can use OptinMonster's testing tool to test social media landing pages. Test the landing page’s effectiveness for your next ebook offer before it's published on LinkedIn first. If you want a platform that offers intuitive tools for social testing, try OptinMonster.

3. Optimizely

Price: Case-by-case basis, contact sales for pricing

If you're heavily promoting your website on social media, test it to learn how it performs among your audience. One way you can do that is with Optimizely, which provides marketing software solutions.

Optimzely's tools let you perform split and multivariate tests. Optimizely will count the number of website visitors and the metrics you pick, such as engagement, goals, conversions, and clicks. You'll be able to apply custom audience segments that can be filtered based on social proof performance.

4. Leadpages

Price: Free, or $48-$199/mo.

If you want a tool specifically for testing landing pages you promote on social media, try Leadpages. This is software that was made just for building professional landing pages. You can use the split testing feature with the tool.

When you run split tests with Leadpages, you'll be able to access its analytics the minute it aggregates traffic and engagements. At the conclusion, you'll get a report that includes the total and unique visits, conversions and total conversion rate. Your report will also include specific insights about test variations.

Keep in mind that you can't set a time frame for your tests. From the analytics dashboard, you'll be able to see how many days your test has been running. When you're ready for the test to be complete, you'll have to end it manually. 

5. Inspectlet

Price: Free, or $39-$499/mo.

Inspectlet was created just for testing. It's a software that lets you test multiple different content types, including web pages. You can also create content variations for future social tests, like edited graphics.

The visual editor in Inspectlet makes it easy to code or create different pages. You can change your page's design or code in real-time and access analytics as soon as the test begins. Inspectlet lets you track views, page elements, unique visitors, and custom events.

When you access your performance report, you'll get the metrics based on the goals you set. This can be anything from tracking clicks, user engagement, custom events, or URL views. Basically, anything that you can track with the software can be measured.

Now, you have a couple of options separate from social media sites that you can social test with. With this in mind, you're ready to run the test.

Think back to the example in the beginning of the post. Knowing all you do about social testing now, you're fully equipped to run a video engagement experiment of your own. Will it be on a social channel or with testing software?

One of the many great things about social testing is how creative they can be — you can social test just about any piece of content, and social media sites are starting to make that process easy. Testing your social ads ensures you're providing the most value to your customers, and now, you can.

Don't forget to share this post!

Related articles.

How The Hustle Got 43,876 More Clicks

How to Do A/B Testing: 15 Steps for the Perfect Split Test

What Most Brands Miss With User Testing (That Costs Them Conversions)

Multivariate Testing: How It Differs From A/B Testing

How to A/B Test Your Pricing (And Why It Might Be a Bad Idea)

11 A/B Testing Examples From Real Businesses

15 of the Best A/B Testing Tools for 2024

How to Determine Your A/B Testing Sample Size & Time Frame

These 20 A/B Testing Variables Measure Successful Marketing Campaigns

How to Understand & Calculate Statistical Significance [Example]

Learn more about A/B and how to run better tests.

Marketing software that helps you drive revenue, save time and resources, and measure and optimize your investments — all on one easy-to-use platform

• SUGGESTED TOPICS
• The Magazine
• Newsletters
• Managing Yourself
• Managing Teams
• Work-life Balance
• The Big Idea
• Data & Visuals
• Reading Lists
• Case Selections
• HBR Learning
• Topic Feeds
• Account Settings
• Email Preferences

I Ran 4 Experiments to Break My Social Media Addiction. Here’s What Worked.

• Sarah K. Peck

Track and schedule your usage.

Are you spending too much time on social media? If you’d like to break the habit, you can try a few different techniques. One would be to quit cold turkey for a full month. If that sounds too extreme, you can avoid social media at certain times, like after dinner or before breakfast. Blocker tools like Freedom can help you stay on track. A third approach is to try a social “happy hour” — instead of staying off social media at certain times, block out a portion of every day you can look forward to indulging in it. A fourth experiment to try is a taking a day off from social every week, like a Saturday or Sunday. This “day of rest” will help you keep your social habit in check, and make the weekend feel longer.

Social media can connect us to new ideas, help us share our work, and allow previously unheard voices to influence culture. Yet it can also be a highly addictive time-sink if we’re not careful about our goals , purpose , and usage.

• SP Sarah K. Peck is an author and startup advisor based in New York City. She’s the founder and executive director of Startup Pregnant, a media company documenting the stories of women’s leadership across work and family, and host of the Startup Pregnant Podcast .

Partner Center

More From Forbes

Why ‘the social dilemma’ on netflix is such an important film.

• Share to Facebook
• Share to Twitter
• Share to Linkedin

There’s a girl staring into a mirror.

She has a bland expression on her face, but something's not quite right.

She keeps touching her hair as if she wants to push her ears in. Earlier that day, someone shared an elephant emoji on one of her social media posts. “Can your ears be any bigger?” the person asked in a comment. Maybe it didn’t register at first.

It does now. Looking into the mirror, she pushes her ear in again as a tear drops down her face. You feel the pain she’s experiencing from what was probably meant as a joke.

Sadly, it’s not a joke. This is a scene from the documentary film called The Social Dilemma on Netflix and it’s one of the most important movies the company has ever released, especially if you have kids. When I watched it, I took furious notes about who was speaking and what they said. That’s partly because I’m currently working on a book about good and bad habits, but mostly because it’s a riveting expose.

Look, I know how this all works. Watching the movie, I couldn’t help but think how all of this constant clicking, swiping, and liking has impacted our lives. I’ve covered social media since about 2008 when these apps debuted on mobile devices. 

I wrote about the dangers even back then. My own journey had a major false start. I created a Twitter account in about 2007 and posted a few article links, then noticed someone kept bashing my work over and over. I deleted that Twitter account and started over about six months later. It’s one reason my only option was @jmbrandonbb and not something a bit easier to find. Honestly, I didn’t want my full, real name to appear in the username after that initial foray.

Netflix’s Best New Movie Arrives With A Perfect 100% Critic Score

Trump reposts ai-generated images claiming taylor swift fans support him, microsoft issues mandatory 2fa login deadline alert.

I’m quite a bit older than the middle schooler who appears in that documentary, and that scene is a re-creation that runs throughout the film as a way to depict what life is like now. Some people seem to hate it . To me, it’s spot on. The girl is an actor but I couldn’t help but wonder if the tears were genuine and based on real experiences. We’re all human. In one segment, the brilliant ethicist Tristan Harris notes that we were not really meant to receive feedback on what we do and say every five minutes.

The documentary is exposing some hard truths. It uses terms like surveillance capitalism and positive intermittent reinforcement that, quite honestly, if you don’t already know what they mean then you might already be lulled into the honeytrap. You might already be stuck in one.

“Social media isn’t a tool that’s just waiting to be used,” says Harris. “It has its own goals and it has its own means of pursuing them by using your own psychology against you.”

In other words: The tool is alive. It knows you. It’s feeding you information you think you want and need but in reality is eliciting action and clicks as a way to fuel advertising.

The documentary does not soften any punches. It refers to us as lab rats in a way that is not meant to be funny anymore. We’re all rats at this point. We think it’s all about getting cheese as a reward and it’s harmless, but there is a lot more at stake. Not sure if you know this, but most lab rats don’t live a long and fruitful life.

I learned several other terms. Dopamine deficit state. A pleasure-pain balance. It’s like I’m being pulled out of the matrix, although I’ve known about the allure for some time. Most of us like the matrix, mostly because of the juicy steak .

Tools that are alive are the most dangerous. My view is that it’s time to start viewing digital media and other apps as part of the vast experiment that the movie is describing and to do something about it. There are quite a few methods to combat the allure of social media, but there is one you can do right now after reading this.

We know we’re clicking too much. We know it’s captivating. One simple step to consider: Grab your phone right now and look over your apps. Which one is capturing all of your attention right now? For the techies out there, you can actually find out. On an iPhone, for example, long-press on Settings, go to Battery, and scroll down to see your most used apps. This is not easy to admit, but lately mine has been Instagram. I’ve been documenting the book-writing process but I’m going to delete it for a month.

I’ll report back later on how this all went. If you deleted an app you have been using too much, that has been pulling you away from real life, drop me an email and explain why you use it so much. Let me know if you really did delete it, and what you hope to gain from this exercise. Let’s compare notes on our findings.

• Editorial Standards
• Reprints & Permissions

Every print subscription comes with full digital access

Science News

Social media harms teens’ mental health, mounting evidence shows. what now.

Understanding what is going on in teens’ minds is necessary for targeted policy suggestions

Most teens use social media, often for hours on end. Some social scientists are confident that such use is harming their mental health. Now they want to pinpoint what explains the link.

Carol Yepes/Getty Images

Share this:

By Sujata Gupta

February 20, 2024 at 7:30 am

In January, Mark Zuckerberg, CEO of Facebook’s parent company Meta, appeared at a congressional hearing to answer questions about how social media potentially harms children. Zuckerberg opened by saying: “The existing body of scientific work has not shown a causal link between using social media and young people having worse mental health.”

But many social scientists would disagree with that statement. In recent years, studies have started to show a causal link between teen social media use and reduced well-being or mood disorders, chiefly depression and anxiety.

Ironically, one of the most cited studies into this link focused on Facebook.

Researchers delved into whether the platform’s introduction across college campuses in the mid 2000s increased symptoms associated with depression and anxiety. The answer was a clear yes , says MIT economist Alexey Makarin, a coauthor of the study, which appeared in the November 2022 American Economic Review . “There is still a lot to be explored,” Makarin says, but “[to say] there is no causal evidence that social media causes mental health issues, to that I definitely object.”

The concern, and the studies, come from statistics showing that social media use in teens ages 13 to 17 is now almost ubiquitous. Two-thirds of teens report using TikTok, and some 60 percent of teens report using Instagram or Snapchat, a 2022 survey found. (Only 30 percent said they used Facebook.) Another survey showed that girls, on average, allot roughly 3.4 hours per day to TikTok, Instagram and Facebook, compared with roughly 2.1 hours among boys. At the same time, more teens are showing signs of depression than ever, especially girls ( SN: 6/30/23 ).

As more studies show a strong link between these phenomena, some researchers are starting to shift their attention to possible mechanisms. Why does social media use seem to trigger mental health problems? Why are those effects unevenly distributed among different groups, such as girls or young adults? And can the positives of social media be teased out from the negatives to provide more targeted guidance to teens, their caregivers and policymakers?

“You can’t design good public policy if you don’t know why things are happening,” says Scott Cunningham, an economist at Baylor University in Waco, Texas.

Increasing rigor

Concerns over the effects of social media use in children have been circulating for years, resulting in a massive body of scientific literature. But those mostly correlational studies could not show if teen social media use was harming mental health or if teens with mental health problems were using more social media.

Moreover, the findings from such studies were often inconclusive, or the effects on mental health so small as to be inconsequential. In one study that received considerable media attention, psychologists Amy Orben and Andrew Przybylski combined data from three surveys to see if they could find a link between technology use, including social media, and reduced well-being. The duo gauged the well-being of over 355,000 teenagers by focusing on questions around depression, suicidal thinking and self-esteem.

Digital technology use was associated with a slight decrease in adolescent well-being , Orben, now of the University of Cambridge, and Przybylski, of the University of Oxford, reported in 2019 in Nature Human Behaviour . But the duo downplayed that finding, noting that researchers have observed similar drops in adolescent well-being associated with drinking milk, going to the movies or eating potatoes.

Holes have begun to appear in that narrative thanks to newer, more rigorous studies.

In one longitudinal study, researchers — including Orben and Przybylski — used survey data on social media use and well-being from over 17,400 teens and young adults to look at how individuals’ responses to a question gauging life satisfaction changed between 2011 and 2018. And they dug into how the responses varied by gender, age and time spent on social media.

Social media use was associated with a drop in well-being among teens during certain developmental periods, chiefly puberty and young adulthood, the team reported in 2022 in Nature Communications . That translated to lower well-being scores around ages 11 to 13 for girls and ages 14 to 15 for boys. Both groups also reported a drop in well-being around age 19. Moreover, among the older teens, the team found evidence for the Goldilocks Hypothesis: the idea that both too much and too little time spent on social media can harm mental health.

“There’s hardly any effect if you look over everybody. But if you look at specific age groups, at particularly what [Orben] calls ‘windows of sensitivity’ … you see these clear effects,” says L.J. Shrum, a consumer psychologist at HEC Paris who was not involved with this research. His review of studies related to teen social media use and mental health is forthcoming in the Journal of the Association for Consumer Research.

Cause and effect

That longitudinal study hints at causation, researchers say. But one of the clearest ways to pin down cause and effect is through natural or quasi-experiments. For these in-the-wild experiments, researchers must identify situations where the rollout of a societal “treatment” is staggered across space and time. They can then compare outcomes among members of the group who received the treatment to those still in the queue — the control group.

That was the approach Makarin and his team used in their study of Facebook. The researchers homed in on the staggered rollout of Facebook across 775 college campuses from 2004 to 2006. They combined that rollout data with student responses to the National College Health Assessment, a widely used survey of college students’ mental and physical health.

The team then sought to understand if those survey questions captured diagnosable mental health problems. Specifically, they had roughly 500 undergraduate students respond to questions both in the National College Health Assessment and in validated screening tools for depression and anxiety. They found that mental health scores on the assessment predicted scores on the screenings. That suggested that a drop in well-being on the college survey was a good proxy for a corresponding increase in diagnosable mental health disorders. 

Compared with campuses that had not yet gained access to Facebook, college campuses with Facebook experienced a 2 percentage point increase in the number of students who met the diagnostic criteria for anxiety or depression, the team found.

When it comes to showing a causal link between social media use in teens and worse mental health, “that study really is the crown jewel right now,” says Cunningham, who was not involved in that research.

A need for nuance

The social media landscape today is vastly different than the landscape of 20 years ago. Facebook is now optimized for maximum addiction, Shrum says, and other newer platforms, such as Snapchat, Instagram and TikTok, have since copied and built on those features. Paired with the ubiquity of social media in general, the negative effects on mental health may well be larger now.

Moreover, social media research tends to focus on young adults — an easier cohort to study than minors. That needs to change, Cunningham says. “Most of us are worried about our high school kids and younger.” 

And so, researchers must pivot accordingly. Crucially, simple comparisons of social media users and nonusers no longer make sense. As Orben and Przybylski’s 2022 work suggested, a teen not on social media might well feel worse than one who briefly logs on. 

Researchers must also dig into why, and under what circumstances, social media use can harm mental health, Cunningham says. Explanations for this link abound. For instance, social media is thought to crowd out other activities or increase people’s likelihood of comparing themselves unfavorably with others. But big data studies, with their reliance on existing surveys and statistical analyses, cannot address those deeper questions. “These kinds of papers, there’s nothing you can really ask … to find these plausible mechanisms,” Cunningham says.

One ongoing effort to understand social media use from this more nuanced vantage point is the SMART Schools project out of the University of Birmingham in England. Pedagogical expert Victoria Goodyear and her team are comparing mental and physical health outcomes among children who attend schools that have restricted cell phone use to those attending schools without such a policy. The researchers described the protocol of that study of 30 schools and over 1,000 students in the July BMJ Open.

Goodyear and colleagues are also combining that natural experiment with qualitative research. They met with 36 five-person focus groups each consisting of all students, all parents or all educators at six of those schools. The team hopes to learn how students use their phones during the day, how usage practices make students feel, and what the various parties think of restrictions on cell phone use during the school day.

Talking to teens and those in their orbit is the best way to get at the mechanisms by which social media influences well-being — for better or worse, Goodyear says. Moving beyond big data to this more personal approach, however, takes considerable time and effort. “Social media has increased in pace and momentum very, very quickly,” she says. “And research takes a long time to catch up with that process.”

Until that catch-up occurs, though, researchers cannot dole out much advice. “What guidance could we provide to young people, parents and schools to help maintain the positives of social media use?” Goodyear asks. “There’s not concrete evidence yet.”

More Stories from Science News on Science & Society

Old books can have unsafe levels of chromium, but readers’ risk is low

Astronauts actually get stuck in space all the time

Scientists are getting serious about UFOs. Here’s why

‘Then I Am Myself the World’ ponders what it means to be conscious

Twisters asks if you can 'tame' a tornado. We have the answer

The world has water problems. This book has solutions

Does social status shape height?

In ‘Warming Up,’ the sports world’s newest opponent is climate change

Subscribers, enter your e-mail address for full access to the Science News archives and digital editions.

Not a subscriber? Become one now .

How to run and measure social media experiments

You already know that social moves fast. What worked for your brand a few months ago may not be relevant today. This is why social media managers thrive when they embrace a mindset of continual learning and development. Improving your social media marketing strategy requires frequent reevaluation and iteration, and running social media experiments is an essential part of the process.

Whenever you have a hypothesis, question or challenge related to your social media marketing strategy, social media experiments can provide actionable next steps. Their results provide concrete evidence to support your case for more resources or reasoning behind switching up your current content.

Social media experiments not only challenge your current strategy, but can also open opportunities to try something different—such as a new social media network or feature—and determine if it’s effective for your target audience. Experimentation can also reveal faster ways to reach your goals, help you avoid costly mistakes and uncover new information about your audience.

Grab your metaphorical safety goggles, lab coat and test tubes because in this article we’re going to walk through the steps for running and measuring successful social media experiments.

7 Steps for running a social media experiment

With these seven steps, you’ll be testing on social media with ease in no time:

• Formulate a hypothesis
• Choose the right type of social media experiment
• Select your metrics and the network you want to test
• Define the duration of the social media experiment
• Select your variables and control
• Conduct the social media experiment
• Analyze and share the results of your experiment

 1. Formulate a hypothesis

Before you begin, you’ll need a basic understanding of the following:

• The overall goals of your business
• Your current social strategy, including overarching goals per platform
• Your audiences by social network
• Your current social performance
• The questions, notions and ideas you wish to test

Prioritize a hypothesis that will result in the biggest impact on your team’s top-level social media goals . Avoid running several tests at once because it can lead to inconclusive results, especially if you’re focused on managing organic social.

If you’re using Sprout, you can learn about your audiences and performance by channel through our cross-network reports (like the Post Performance Report) or competitor reports (like the Instagram Competitors Report).

To dive even deeper into understanding your audience, use Sprout’s Advanced Listening tools. With Listening, you can build queries to track and analyze social conversations, pin down trends and view consumer sentiments. Seeing the data behind what your audience is talking about and the content they engage with will help you formulate a hypothesis.

2. Choose the right type of social media experiment

Now that you have a hypothesis, it’s time to select the type of social media experiment you will conduct to prove your theory.

There are two main types you can choose from: A/B testing and multivariable testing.

Social media experiment ideas for A/B tests

One of the most common types of social media experiments, an A/B test is an experiment where you change only one variable and keep everything else the same. These types of tests are an excellent way to pinpoint improvements that will make a measurable impact. Some common A/B tests on social include:

• Content types: video vs. a link, photo, GIF, etc.
• Captions: long vs. short
• Copy: question vs. statement, emojis or hashtags
• Images: illustrations vs. photography or animation
• Posting time: Monday at 9:00 a.m. vs. Friday at 4:00 p.m.

For example, if you wanted to test which content type is the most engaging on Instagram Stories, your team could test photo content against video content. The content type would change, but you would use the same caption and post at the same time and day of the week, one week apart.

Using Sprout, the Atlanta Hawks ‘ social team tested a casual approach to videos at community events. A player shot a hand-held video that was compared to the performance of more produced social videos. The casual video format proved to be more successful and sharing the performance data was a major win for the social team.

Social media experiment ideas for multivariable testing

As its name implies, multivariable testing alters two or three variables at once. However, since you’re experimenting with more elements, analyzing and interpreting data can be harder. You’ll also need a large audience to avoid skewing the test.

Some multivariable tests include:

• Short-form animated video vs. long-form live action video
• Varying tones of voice paired with or without emojis
• Multiple call-to-action buttons with different featured images
• Different content types with various captions
• Same content type but different days/times and platforms to see which resonates the most, like Instagram vs. TikTok

Sprout’s social team conducted several multivariable tests to help develop our TikTok marketing strategy , as you’re about to read in the next step.

3.  Select your metrics and the network you want to test

Establish the key metric you want to measure successful content against. This can include impressions, traffic to a particular page such as your brand’s website or a gated resource, and engagement metrics (Think: likes, clicks, comments or shares).

The channel you choose to conduct your experiment will depend on what you’re testing and the social media network you use the most to post that kind of content. Use your network-specific data to inform this decision. Read some of Sprout’s Insights resources to learn which content types perform the best on which platforms.

When our social team started testing TikTok, the main goal was to increase awareness among our target audiences. Accordingly, we selected impressions, video views, profile views and audience growth as key performance indicators.

4. Define the duration of the social media experiment

Don’t fall into the common mistake of not defining a time frame for your social media experiment. Remember that social media strategy is a long game–give time for new initiatives to grow and develop.

Your reporting window depends on your budget, audience size and KPIs, but the most important factor is to reach statistical significance.

Statistical significance refers to the likelihood your test results are the outcome of a defined cause and not chance. To reach statistical significance, you’ll need a large sample size and a control. For example, a sample size of 1,000 is stronger than 100, and your control would be the piece of content you do not change.

Set a duration and look for statistical significance. What are the significance changes? After your testing period, consider optimizing content that didn’t work during that timeframe instead of hitting the breaks on posts that aren’t resonating immediately.

While experimenting with TikTok, the social team reported results after four months since there was enough data available to analyze. They also set a weekly update to our internal social dashboard to continue testing and learning, along with iterating strategy, if needed.

During the first four months, we discovered views for every TikTok remained consistent, with an average of 535 views per video. We were also able to confirm our thoughts/assumptions about the For You Page (FYP) and the TikTok algorithm—each consistently pushed out content to our target audience (social media specialists, managers, digital marketers, etc.).

5. Select your variables and control

If you’re using A/B testing, consider all of the elements of your content that could influence your test results to ensure you’re only testing one variable. Also select your control, which is the content that will not change. For example, if you’re testing images, make sure to not change the copy, audience, timing, etc.

In our social team’s multivariable TikTok experiments, they tested several variables including formats, themes and creative considerations like music, sounds and closed captions.

In the example below, 91% of views came from the FYP, 5% came from a personal profile view and 1% came from direct followers–confirming their hypothesis that the FYP and the algorithm were the key drivers pushing out content to our target audience.

If you use Sprout, you can use tagging to track the performance of your control and the test post.

6. Conduct the social media experiment

Now it’s time to execute! Use Sprout’s Publishing tools to seamlessly plan, create, optimize and post your content for the experiment. For example, you can use Sprout’s ViralPost® technology to post at optimal send times.

Use the Tag Performance Report to organize, run and analyze your social media experiment results, including your paid campaigns.

Read our guide on creative testing for more tips and examples for conducting social media experiments.

7. Analyze and share the results of your experiment

Review the results of your experiment to identify new opportunities or add insights to your records.

If you’re trying to gain executive buy-in, especially for further testing or resources, you’ll need to communicate and create an effective data story to highlight why your company will benefit from your suggested next steps.

Using Sprout, you can easily access automated, presentation-ready reports to help illustrate your data story. Create custom reports, like this Facebook Performance Summary that includes impressions, engagements, post link clicks and publishing behavior for various content types:

Use experiments to optimize engagement and growth

Here’s a quick overview of the seven steps:

Good luck on your journey to embracing curiosity and thinking like a scientist—your social strategy will thank you.

This article is an excellent first step, but there’s so much more to learn about social media experiments. Step into the (virtual) lab yourself and get a hands-on experience, by signing up for a free trial .

Related Posts

Let’s cut to the chase: It’s not organic vs. paid social media—it’s organic and paid social media.  As social becomes…

As a UX designer, you get to work on creative, rewarding, even life-changing projects. It’s an industry with flexible working…

Welcome to our first Creative Essentials post, a new series that focuses on the tools and resources used by different…

Micro-influencer marketing can be an excellent strategy for businesses with smaller marketing budgets. And because micro-influencers take up 91% of…

As we head into the final month of 2022, plenty of new ideas and website design trends are still emerging….

A few days ago, the big tech news was that Jack Dorsey was stepping down from his role as Twitter…

Tracking social media engagement across a bunch of different networks can be tricky. But those likes, comments and shares are…

LinkedIn is more than just a networking site. It’s a top spot for generating leads and getting your brand seen….

Trying to figure out how to make money on YouTube? Because there’s a big window of opportunity for monetizing videos….

LinkedIn ads are a powerful tool for all kinds of businesses. The platform has over 810 million users, with 65…

If you’re building a content marketing strategy, you’re probably looking at blogging first. Blogs give you the opportunity to go…

If you can believe it, Sprout recently celebrated its 14th anniversary. Thinking back on what social media looked like in…

There’s nothing better than the “aha!” moment when you realize there’s a better way to manage your brand accounts. Even…

Our good friend Cody Ellingham has shared with us his latest project titled: BANGKOK PHOSPHORS, it’s basically a photobook of…

For me to use the term “branding industry” often feels odd, even though I work within it. The entry point…

Scientists Studied People Who Don't Use Any Social Media. Here's What They Learned

Many of us spend hours every day tethered to our devices, pawing at the screen to see if it will deliver a few more likes or emails, monitoring the world and honing our online presence.

Social networking platforms such as Whatsapp, Snapchat, Instagram, Facebook and Twitter are supposed to make us feel more connected. Yet our reliance on technology to "see" the social world around us can be a heavy burden.

The Pew Research Centre recently reported that about a quarter of US adults say they are "almost constantly" online.

Stress, addiction, depression and anxiety seem unsurprising consequences of using social platforms often specifically designed to keep us repeating the same actions over and over again.

Even so, many would find the prospect of living offline worrisome, or simply impossible. That's why we undertook a small study with 50 people who may seem nothing less than social outcasts in today's screen saturated environment.

None of our participants used social media or had a mobile phone, and most even refused to email.

We wanted to understand why these people had decided to switch off, and how they managed it. But rather than seeking quick fixes for overuse, we explored the principles and values that drove our participants to live the way they do.

Much has already been written about how we can switch off – but that won't achieve much, unless we really feel the benefits.

Here's what our respondents said they'd learned, from living their social lives offline.

1. Spending time with others

Part of the problem with social networking platforms is that we don't just use them for communicating – they also promote a particular way of being connected to and supportive of those around us.

These interactions are channelled through the platform to create data, which is ultimately fed back to data brokers and marketers.

Our participants shared a deep belief in, and attachment to, a different way of socialising that's focused on expression, touching, talking and being in the same space, physically.

For them, this helped to maintain a feeling of human bonding and connection.

And while this slower, deeper acknowledgement of others was especially valued by our participants, they also thought it might be valuable to society more broadly.

Given the angst-ridden nature of frenetic social networking, we could all benefit from slowing down and taking stock more often.

For many people today, the sense of being "always-on" is generating a desire to achieve greater balance and disengage from the things that are causing them stress.

For our participants, who didn't use smartphones and social media, time with others was associated with a sense of calm and purpose in life.

2. Switching off is not missing out

Our participants questioned what exactly is "social" about social media: what constitutes communication, and what do we get from the way that social stuff is measured on online platforms – whether that's friendship, support or social contact.

Rather than having hundreds of "friends", they would always choose to see people face to face and nurture relationships that would support them through the tough times.

Taking the opportunity to switch off may, at first, cause some anxiety. But the trick is to realise that switching off is not the same as missing out. When you first switch off, you may spend more time in your own company.

But from these moments may come a realisation of how exhausting it is to sustain online connections, and indeed how superficial it is to be locked in endless exchanges of trivial information.

Those who chose to disconnect are neither sad nor excluded. Freed from the screen, they escaped from the overwhelming flows of information and tasks. Their deep sense of connection with the world, and their loved ones, was clear to see.

3. Being, rather than doing

Many of those who switched off enjoyed new-found vitality, because they found time to connect with the world in the here and now. This is crucial to helping us reset and relax , so that we are prepared for more stressful times.

Time spent scrolling through content may feel as though it makes light demands on body and mind. But the visual interference from a bright screen is far from relaxing.

You are much less likely to have restful sleep if you share a bed with your smartphone, or surf to sleep .

As mindfulness is becoming more popular, its core ideas are often coopted by technology. On Instagram, for example, successful influencers show off their yoga skills and promote spiritual disciplines.

Fitness trackers, health data and yoga apps consistently rank among the top apps downloaded by smartphone users.

Our disconnected group told us that we should be more critical of our use of apps and start leaving our phone behind.

If mindfulness is a state of being focusing on the present – channelling thoughts, feelings and sensations as they flow through us – then what use is a screen?

Constant connection paradoxically results in less free time, and periods when we are able to think without interruption give precious refuge from the demands of daily life.

These disconnected people did not switch off to be "anti-social". They did so to take charge of when and where they connected with people. They may well be part of a vanguard, leading to new ways of being happier, more rested and, yes, more social.

Ten years from now, we might look back at the emergence of social media as a part of humanity's growing-up – a time that created social divisions, anxiety and restlessness and which damaged the health and well-being of many.

Rowland Atkinson , Chair in Inclusive Societies, University of Sheffield and Mariann Hardey , Directorate Advanced Research Computing (ARC) Durham University, Durham University

This article is republished from The Conversation under a Creative Commons license. Read the original article .

• Social Media Marketing

A Simple 6-Step Framework for Running Social Media Experiments (with 87 Ideas Included)

Product Marketing @ Buffer

Experiment with ideas. Test and see which works better. Analyze your data.

These are phrases we often use on this blog. To us, social media marketing is a bit of a science . We recommend testing things, running experiments, and analyzing data — because it worked for us. This experimental mindset has helped us grow our social media results .

But one thing we haven’t done well is to explain the how: how to run social media experiments .

In this post, you’ll learn the six simple steps of running social media experiments. We’ve even included 87 ideas, which you can start testing immediately.

How to run social media experiments successfully

Running social media experiments can be hard when you’re not sure where to start and where to head to. Here’s a step-by-step guide to help you hit the ground running.

Before we dive into the guide, here’s a quick caveat: social media experiments are not perfect or entirely scientific . Some factors are out of our control, such as organic reach since it’s determined by the social media algorithms .

This doesn’t mean we wouldn’t get meaningful results (it has worked for us and many others); it’s just good to be mindful of this while running your experiments.

1. Set goals

As with most planning, it’s crucial to start by setting your goals. Why?

Imagine the following situation. Both social media posts are sharing the same blog post with a different headline.

Post A received 100 Likes, 100 shares, 10 clicks, and 5,000 impressions.

Post B received 10 Likes, 10 shares, 100 clicks, and 1,000 impressions.

Which post do you think is better?

I think it depends on your goals! If you think social media is for engagement , you’ll likely prefer Post A. But if you think social media is for driving traffic, you’ll probably prefer Post B instead.

Here’s a list of social media goals you could choose from:

• Reach (or impressions)
• Engagement (Likes, comments, and shares)
• Traffic from social media
• Leads from social media
• Revenue from social media

For us, our overarching goal for social media is engagement and brand-building. ( Here’s why .) So we focus more on our social media reach, engagement, and following than traffic, leads, or revenue from social media.

Having said that, each social media post can sometimes have its own micro-goal. For example, while our overall social media goal is engagement and majority of our posts are meant for generating engagement, we have some posts that are meant for driving traffic, such as this and this .

2. Brainstorm ideas

Once you have set your goals, you are ready to come up with ideas. While you are thinking of new ideas, it’ll be good if you could form a hypothesis around the idea, too. This is the format we like to use

If we  (experiment idea), then (expected results), because (assumptions).

If we curate top content from other Facebook Pages, then we can grow our Facebook reach by 10%, because they are content proven to be popular.

You could also keep it as simple as “Curating top third-party content will increase our reach on Facebook.”

Here are a few suggestions for coming up with social media experiment ideas:

Read blog posts for ideas

This is my favorite method because there’s so much written about social media marketing every day. Listicles and case studies of successful social media tactics can be a great source of inspiration for experiment ideas.

If you want somewhere to get started, we have quite a few blog posts with experiment ideas in them:

• Get Over Your Creativity Block With These 20 Social Media Content Ideas
• 20 Creative Ways to Use Social Media for Storytelling
• 7 Powerful Social Media Experiments That Grew Our Traffic by 241% in 8 Months
• 10 Unique Ideas to Test on Every Social Media Channel (And How to Tell What Works)
• We Made These 10 Social Media Mistakes so You Don’t Have To

There’s also a huge list of 100 social media experiment ideas below. Click here to skip right to it , and feel free to take any of the ideas.

Follow social media trends

The second method is to follow social media trends .

For example, videos are becoming the most popular content format on social media. Facebook has been pushing for videos on its platform for the last few years, and LinkedIn has recently introduced native videos. Our internal data also showed that videos received an average of 873 interactions per post , compared with 279 for photos and 190 for text posts.

So it’ll be a good idea to test videos on your social media profiles .

We recently wrote about the 10 major social media trends for 2018 , which you might find useful for generating ideas.

What ideas can you think of in light of these trends?

Study industry leaders and competitors

The final method is to watch and learn from the best companies in your industry and your competitors. What have they been doing that is worth trying yourself?

It’s also good to be aware that the ideas that worked for them might not always work for you. You all likely have many differences such as branding, positioning, and audience. But if you think an idea is suitable for your brand, I would say go for it and modify it for your own brand.

There are some free tools you can use to track industry leaders and competitors.

On Facebook, you have Pages to Watch . It allows you to quickly check out the recent top posts of similar Facebook Pages. You can find it at the bottom of the Overview tab in your Page Insights .

On Twitter, you could add your favorite companies to a Twitter list . To create a Twitter list, click on your profile photo in the upper-right corner and click on “Lists”. Then, click on “Create new list” and fill out the information.

On LinkedIn, you have the Companies to track feature in your Company Page analytics. Clicking on any of the company names will bring you to their Company Page. You can access this section by clicking on Analytics > Followers.

3. Prioritize

The next step is to prioritize your ideas. A prioritization framework we like to use at Buffer is the ICE score by GrowthHackers .

ICE stands for Impact, Confidence, and Ease.

• Impact: The possible impact of the idea on your selected metric (e.g. 10 percent increase in reach)
• Confidence: Your confidence level about the success of the experiment (e.g. three companies have found success with this idea)
• Ease: The number of resources required (e.g. no design or engineering help needed)

For each experiment, give each factor a score from one to 10. The overall score is determined by taking the average of the three scores. You should start with the experiment that has the highest score.

Here are two simple examples of ICE scoring in action:

Experiment A Curating top third-party content will increase our reach on Facebook. Impact = 6 Confidence = 8 Ease = 8 Overall = 7.3

Experiment B Partnering with micro-influencers will grow our Instagram reach. Impact = 8 Confidence = 4 Ease = 3 Overall = 5

Based on the ICE score, I would run experiment A before experiment B.

While this process can be a little time-consuming at the start, it’s important. It’ll help you think through the experiment details (such as what metric to track) and maximize your impact with the resources you have.

After a while, you should be able to build up a good intuition about the potential of ideas without having to score every single idea.

Now you’re ready to test your top ideas!

There are a few things you want to be mindful when testing.

(Ideally) test one thing at a time  to understand what’s making the difference. For example, if you want to test your copy as an experiment, it’ll be best to keep the multimedia the same. Otherwise, you won’t know if the copy or the multimedia caused one post to outperform the other.

Look at the right metrics  to measure the results of your experiment. This is where the goals you’ve set will be helpful. For instance, if you want to maximize your social media reach, you would pick reach or impressions over clicks.

Run one experiment at a time  for a start. Similar to the first point above, doing so lets you know which experiment moved the needle. (When you are feeling advanced, you could run multiple experiments concurrently as long as you understand how they will affect the metrics.)

Run each experiment for at least a week  for smaller experiments. This isn’t entirely scientific but I believe a week is sufficient for the results to be seen. For bigger experiments such as shifting your social strategy to posting more videos, you might want to test it for a month to a quarter. The bigger the experiment, the longer it should be tested.

5. Analyze and learn

Finally, you’ll want to analyze your results to see if your experiment has been a success. Here are some questions you could ask yourself:

• Did it achieve the results I had expected? Why?
• Did any other factors contribute to the success or failure?
• Can I learn anything else from this experiment?

To help you with your experiment tracking, I’ve created a simple tracking template: Social Media Experiment Tracking . Feel free to make a copy and modify it to your liking.

When running scientific experiments, it’s important to look at the statistical significance of the results — to ensure that the result isn’t a fluke and can be repeated successfully. But for social media experiments, it might not always be feasible. That’s because your sample size (impressions of a post) isn’t within your control.

My non-scientific recommendation here is to repeat the experiment a few times and see if the result remains the same . If the result can be repeated, you can consider turning the experiment into a regular part of your social media marketing.

Congratulations! You have just planned, run, and analyze a social media experiment!

Whether you had a successful experiment or not, it’ll be great to repeat step four (test) and five (analyze and learn) continuously. Then once a quarter, you could take a step back and look at the bigger picture again . Your social media goals might have changed, or there might be new social media tactics to try.

Re-evaluate your goals, brainstorm new ideas, and test them. All the best!

87 social media experiment ideas

To help you get started with running social media experiments, here’s a mega list of ideas for you to try. Some are low-hanging fruits while others might require much effort:

Posting time

• Post when your followers are online
• Post when your followers are offline
• Post during commute times
• Post during lunch time
• Post on the weekends

Posting frequency

Headlines and copy.

• Write short headlines
• Write long headlines
• Write really long headlines (or stories )
• Use social proof in your copy
• Customize your post for each social media platform
• Post questions
• Ask for opinions on a trending topic
• Share top industry news
• Share thought-leadership articles
• Share interesting, relevant statistics
• Share inspiration quotes
• Post interviews
• Host a live Q&A
• Post behind-the-scenes videos
• Share your company culture
• Re-use top posts
• Poll your audience
• Retweet a mention every day
• Create a branded hashtag
• Host a giveaway and invite people to comment
• Host a giveaway and invite people to tag a friend
• Host a giveaway and invite people to share your post
• Host a giveaway with other brands
• Celebrate national or international events
• Create a huge image on your Instagram profile with multiple posts
• Create a Twitter moment
• Create a Slideshare presentation (and share it)
• Curate third-party content
• Post self-explanatory images
• Post photos of your product
• Post infographics
• Post audio recordings
• Post slideshow videos
• Post tutorial or tips videos
• Post a 360 photo or video
• Livestream an event
• Live-tweet an event
• Upload videos directly to social media platforms (vs YouTube)
• Create landscape videos
• Create square (or letterbox) videos
• Create portrait videos
• Create short 10-15s videos
• Create long 20-30min videos
• Add captions to videos
• Add music to videos
• Use a cover video for Facebook
• Boost your top posts
• Use a photo of a person
• Test the carousel ad format
• Test the video ad format
• Test stories ads
• Test Messenger ads
• Test Snapchat Geofilters
• Test Snap Ads
• Sequence your Facebook ads

Collaboration

• Share user-generated content
• Share customer stories
• Sponsor a micro-influencer (sponsored posts)
• Create a piece of content with a micro-influencer and share it together
• Host a social media takeover
• Do a social swap
• Host social media events with another brand
• Host a roundtable with experts in your industry
• Hire an agency for a social media campaign
• Start a Twitter chat
• Create (and link) a Facebook Group
• Create a LinkedIn Group
• Reply to all mentions
• Use Facebook Messenger
• Create a Facebook Messenger bot
• Use a social media management tool
• Use a social media analytics tool
• Use Facebook’s preferred audience feature
• Offer time-limited discounts
• Ask your CEO (or a colleague that is well-known in your industry) to share your posts

How do you run social media experiments?

Having a framework for running social media experiments can be very helpful. Here’s one I like (though you can tweak it however much you like):

• Analyze and learn

I’m curious about how you run your social media experiments. Do you use a framework or system? Do you use any tools to help you with it? Let’s chat in the comments section below.

The amazing featured image is by chuttersnap , taken from Unsplash .

Try Buffer for free

140,000+ small businesses like yours use Buffer to build their brand on social media every month

Related Articles

Everything you need to know before your first TikTok Live, including a step-by-step guide and best practices, along with some ideas to inspire you.

We walk through 8 ideas for Instagram monetization in this detailed article.

A deep dive into the differences between Buffer and Hootsuite, including pricing, features, and who each social media tool might be the best fit for.

140,000+ people like you use Buffer to build their brand on social media every month

Copyright © 2024 Buffer | Privacy | Terms | Security

• Instagram Marketing
• Social Analytics
• News and Trends
• Best apps and tools
• Inside Buffer
• Remote Work
• Transparency
• Blog & podcasts
• Best Time to Post on Social Media
• Social Media Blog
• Social Media Glossary
• About Buffer
• Our culture

To revisit this article, visit My Profile, then View saved stories .

• The Big Story
• Newsletters
• Steven Levy's Plaintext Column
• WIRED Classics from the Archive
• WIRED Insider
• WIRED Consulting

A Unique Experiment That Could Make Social Media Better

Social media, news, music, shopping, and other sites all rely on recommender systems: algorithms that personalize what each individual user sees. These systems are  largely driven by predictions of what each person will click, like, share, buy, and so on, usually shorthanded as “engagement.” These reactions can contain  useful information about what’s important to us, but—as the existence of clickbait proves—just because we click on it doesn’t mean it’s good. 

Many critics  argue that platforms should not try to maximize engagement, but instead optimize for some measure of long-term  value for users. Some of the people who work for these platforms agree: Meta and other social media platforms, for example, have for some time been working on  incorporating more direct feedback into recommender systems. 

For the past two years, we have been collaborating with Meta employees—as well as researchers from the University of Toronto, UC Berkeley, MIT, Harvard, Stanford, and  KAIST , plus representatives from nonprofits and advocacy organizations—to do research that advances these efforts. This involves an experimental change to Facebook’s feed ranking—for users who choose to participate in our study—in order to make it respond to their feedback over a period of several months. 

Here’s how our study, which launches later this year, will work: Over three months, we will repeatedly ask participants about their experiences on the Facebook feed using a survey that aims to measure positive experiences, including spending time online with friends and getting good advice. (Our survey is a modified version of the previously validated  Online Social Support Scale .) Then we’ll try to model the relationship between what was in a participant’s feed—for example, which sources and topics they saw—and their answers over time. Using this predictive model, we’ll then run the experiment again, this time trying to select the content that we think will lead to the best outcomes over time, as measured by the recurring surveys.

Our goal is to show that it’s technically possible to drive content selection algorithms by asking users about their experiences over a sustained period of time, rather than relying primarily on their immediate online reactions. 

We’re not suggesting that Meta, or any other company, should prioritize the specific survey questions we’re using. There are  many ways to assess the long-term impact and value of recommendations, and there isn’t yet any consensus on which metrics to use or how to balance competing goals. Rather, the goal of this collaboration is to show how, potentially,  any survey measure could be used to drive content recommendations toward chosen long-term outcomes. This might be applied to any recommender system on any platform. While engagement will always be a  key signal , this work will establish both the principle and the technique for incorporating other information, including longer-term consequences. If this works, it might help the entire industry build products that lead to better user experiences.

A study like ours has never been done before, in part due to serious distrust between the researchers studying how to improve recommender systems and the platforms that operate them. Our experience shows just how difficult it is to arrange such an experiment, and how important it is to do so.

The project came out of informal conversations between an independent researcher and a Meta product manager more than two years ago. We then assembled the academic team, as well as  researchers from nonprofits and advocacy groups to help keep the focus on public benefit. Perhaps we were naive, but we were taken aback by rejections from people who nevertheless agreed that we were asking valuable questions. Some organizations passed because of the communications risk, or because some of their staff argued that collaborations with Big Tech are PR efforts at best, if not outright unethical.

Some of the pushback comes from the fact that Meta is putting money toward the project. Although no external researchers are being paid, the University of Toronto has contracted with Meta to manage the university-based parts of the collaboration. This project has significant administrative and engineering costs, in part because we decided to ensure research integrity by externally writing key parts of the code that Meta will run. This funding might have been more trouble than it was worth, but there’s also no reason researchers should have to scrape together pennies or spend taxpayer money when working with the largest companies in the world to develop socially beneficial technology. In the future, third-party funders could support the academic and civil society end of platform research collaborations, as they have sometimes  done .

The problem with instinctive distrust of platforms is not that platforms are above criticism, but that blanket distrust blocks some of the most valuable work that can be done to make these systems less harmful, more beneficial, and more open. Many observers are placing their hopes in transparency, especially transparency required by law. The recently passed EU Digital Services Act  requires platforms to make data available to qualified researchers, and a number of similar policy proposals have been introduced to the US Congress. Yet our work necessarily goes far beyond “data access.”

In our view,  only an experiment that involves intervening on a live platform can test the hypothesis that recommender systems can be oriented to long-term positive outcomes, and develop sharable technology to do so. More than that, it’s unlikely that law alone can compel a company to engage in good faith on a complex project like this one; designing the core experiment took over a year and wouldn’t have been possible without the expertise of the Meta engineers who work with the platform’s technology daily. In any case, attempts to pass  American laws ensuring researcher access to data have, so far, gone nowhere.

Yet collaborative experiments with public results are disincentivized. The answer isn’t to do technosocial research  in secret —or worse, not at all—but to do it to  higher ethical standards . Our experiment is being overseen by the University of Toronto’s human subjects experimentation review process ( IRB ), which is recognized by all the other universities involved as meeting their ethics requirements. All of the users in our study will have given informed consent to participate, and will be paid for their time. We were happy to find champions within Meta who believe in open research.

This level of cooperation requires navigating complex expectations about what information can, should, and won’t be shared. We designed a novel approach to resolving disagreements about confidentiality. We received contractual guarantees that our research will result in a scientific publication meeting peer review standards, and can’t be altered or held up for any reason other than legitimate privacy and confidentiality concerns. We also negotiated the freedom to talk publicly about our collaboration, and in the event the project is halted, the freedom to disclose the reasons why. We’re pretty sure nobody has seen an agreement like this before in an academic-industry collaboration. It took time to design and negotiate this new way of doing research.  

Finally, we insisted that the results be in the public domain, including any resulting intellectual property. We are trying to shift industry norms of secrecy, because virtually every platform faces similar challenges. Everyone would benefit from routine sharing of research.

When we started two years ago, the first reaction to this project was skepticism: “Meta will never do this, and I wouldn’t work with them even if they did.” Today the reaction is more often, “how can we do this too?” It now seems obvious that open research is the only way of addressing the intricate challenges of society-scale algorithms in a democratically legitimate way.

The risks haven’t gone away; we haven’t actually run the experiment yet. Collaborative science moves slower than industry, and Meta’s business priorities and regulatory environment can change quickly. Nor have we yet had to resolve any significant disagreements about what can and cannot be shared publicly. Either party could still derail this project, and set back societally important platform research by years. But we think there’s no substitute for taking such gambles, as researchers cannot run platform experiments alone and platforms cannot achieve legitimacy without openness. There is a crucial place for criticism and accountability, but something more optimistic is also needed to advance the field. We’re all better off when this sort of work happens.

You Might Also Like …

In your inbox: The best and weirdest stories from WIRED’s archive

How the brain decides what to remember

The Big Story: Meet Priscila, queen of the rideshare mafia

Silicon Valley's soulless plutocrats flip for Donald Trump

Event: Join us for The Big Interview on December 3 in San Francisco

Social Media and Job Market Success: A Field Experiment on Twitter

80 Pages Posted: 20 May 2024

University of Michigan, School of Information

University of Michigan at Ann Arbor - School of Information

Alvin E. Roth

National Bureau of Economic Research (NBER); Dept. of Economics, Stanford University

Date Written: May 20, 2024

We conducted a field experiment on Twitter to examine the impact of social media promotion on job market outcomes in economics. Half of the 519 job market papers tweeted from our research account were randomly assigned to be quote-tweeted by prominent economists. Papers assigned to be quote-tweeted received 442% more views and 303% more likes. Moreover, candidates in the treatment group received one additional flyout, with women receiving 0.9 more job offers. These findings suggest that social media promotion can improve the visibility and success of job market candidates, especially for underrepresented groups in economics such as women.

Keywords: social media, academic job market, field experiment

JEL Classification: D47, C78, C92, D82

Suggested Citation: Suggested Citation

University of Michigan, School of Information ( email )

304 West Hall 550 East University Ann Arbor, MI 48109-1092 United States 48109 (Fax)

HOME PAGE: http://jingyiqiu.com

University of Michigan at Ann Arbor - School of Information ( email )

304 West Hall 550 East University Ann Arbor, MI 48109-1092 United States

Alain Cohn (Contact Author)

National bureau of economic research (nber).

1050 Massachusetts Avenue Cambridge, MA 02138 United States

Dept. of Economics, Stanford University ( email )

Landau Economics Building 579 Serra Mall STANFORD, CA 94305-6072 United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics, related ejournals, behavioral & experimental economics ejournal.

Subscribe to this fee journal for more curated articles on this topic

Randomized Social Experiments eJournal

Advertising, marketing & strategic communication ejournal, industrial & organizational psychology ejournal, information policy, ethics, access & use ejournal, psychology research methods ejournal, social media studies ejournal.

Subscribe to this free journal for more curated articles on this topic

Treatment of Imane Khelif and Raygun tells a broader story of the dangers of social media

Topic: Social Media

Social media has made life hell for Imane Khelif and Raygun for very different reasons. ( AP, AAP )

Two incidents — both involving women competing in the Olympics — provide the latest grim proof of the often toxic, menacing and deranged nature of social media.

One was the treatment of Australian Rachael "Raygun" Gunn, who literally broke not just dance moves, but also social media when she became the target of hateful comments from internet trolls after her viral performance during the sport's Olympic debut.

The other, Olympic champion boxer Imane Khelif, was the subject of a worldwide gender eligibility debate and was publicly trolled with vile and inaccurate bigotry across social platforms.

Now JK Rowling and Elon Musk have been named in a cyberbullying lawsuit filed in France by the Algerian boxer. On Wednesday, her attorney Nabil Boudi, said they had filed a criminal complaint over alleged "acts of cyber-harassment" to the Paris public prosecutor's office. The gold medallist said of her lawsuit: "All that is being said about me on social media is immoral. I want to change the minds of people around the world."

Both these women — experts in their fields whether you like them or not — are adults, and therefore you'd hope they'd be more able to deal with the viciousness which has become public discourse, particularly online. But the fact that Raygun had to make a public appeal to be left alone, and Khelif is fighting back through the courts shows that we really are entering a new phase of the social media wars.

This is not to excuse the role of legacy media in this — that would be a mistake. The bullying of them both wasn't isolated to the online world. But it is without a doubt social media where the conspiracies and misinformation about both women grew, festered, and caused acute harm.

Can we clean up social media?

It's not just those in the public spotlight who are on the receiving end of social media nastiness.

The Joint Select Committee on Social Media and Australian Society has been hearing about how social media companies operate in Australia, including the impact they have, particularly on young people, and considering what changes the parliament needs to introduce to deal with what has become a collective crisis. They have been contemplating the big questions of what kind of regulation may be effective in Australia to reduce harm.

It is exploring the decision of Meta to abandon deals under the News Media Bargaining Code, the important role of Australian news and public interest journalism in countering misinformation and disinformation on digital platforms, and broader issues relating to the influences and impacts of what Australians see on social media. An interim report was meant to have landed last week, but it's been delayed because of changes on the committee.

The social media giants have been answering questions raised by politicians on this committee — and their answers show a constant state of defence of their current measures.

For instance, TikTok, well known for its addictive algorithm, says their systems are built around safety. In an answer provided to committee member Liberal MP Zoe McKenzie, TikTok said that for all users under the age of 18, "we have a 60-minute default screen time limit".

"While there's no collectively endorsed position on the 'right' amount of screen time or even the impact of screen time more broadly, we consulted the current academic research and experts from the Digital Wellness Lab at Boston Children's Hospital in choosing this limit," they said.

"For all users, irrespective of age, we make it easy to set screen time limits, so our community can select for themselves a screen time limit that they are comfortable with. We also provide in-app prompts when users have been on the app for a particular period of time, encouraging them to 'take a break'."

The spin is around how fantastic they are already at self-regulating and how they make it easy for individual users to "take a break". But have you tried to take a break on social media? The feedback loop makes it hard and putting the onus on the user alone is laughable.

A ban wouldn't stem the tide

While headlines about an age ban for social media are popular among parents who are bereft of ideas about how to moderate their children's engagement online, there is a growing consensus about forcing accountability of the tech giants to be transparent about the addictive algorithms they create.

In June, Opposition Leader Peter Dutton pledged to ban under-16 year olds from accessing social media in the first 100 days of a Coalition government, blaming the platforms for "a high prevalence of many health conditions".

Prime Minister Anthony Albanese agreed social media is having "a negative impact on young people", pledging $6.5 million to assess age assurance technologies that experts have argued may never work.

The eSafety Commissioner Julie Inman Grant, who joins us on Q+A tonight, has expressed concerns, comparing banning children younger than 16 from social media to banning them from the ocean.

"We don't fence the ocean or keep children entirely out of the water … but we do create protected swimming environments," she says.

eSafety Commissioner Julie Inman Grant. ( Four Corners: Keana Naughton )

In a written submission to the inquiry she warned that the debate around a complete ban for under-16s, enforced by age verification technology, contains a misconception that "social media is a discrete form of media that can be separated from the rest of the internet".

The regulator is worried that if faced with a ban, kids will just access social media in secret. And anyone who has a teenager knows that they are digital natives who laugh out loud at censorship arguments.

This is the concern of independent and committee member Zoe Daniel. She told the ABC law reform needed to reverse the onus onto the company not the user.

"I can understand why parents like it, it's a popular idea that may work in conjunction with other changes but on its own it's a bit of a sideshow," she said.

"It continues to put the onus on the user, which is not where it should be.

"The main game is making the social media companies accountable for the systems that target young people and cause them harm.

"The companies should be made to take responsibility for and be transparent about what they're doing to prevent and mitigate the spread of harmful and toxic content. Rather than targeting the content, we must target the systems."

Daniel told the ABC she is drafting a private member's bill along these lines which she will table later this year.

An experiment

Liberal MP Zoe McKenzie, who sits on the committee, says Australia must tackle the algorithms which feed young people dangerous material.

In a recent speech she revealed that she asked Reset.tech to run an experiment for her, to test how long it would take to get algorithms or systems to recommend Andrew Tate content to someone who was just curious about Jordan Peterson. Reset.tech set up a new 17-year-old male's account on Instagram on a completely fresh handset with a new SIM installed. The Instagram account was clean and there were no previous interactions or activity.

The new account watched and liked 50 of the top Peterson posts and then scrolled through the content being recommended on Instagram reels while continuing to watch and like posts containing Jordan Peterson. 

"After 70 pieces of content, the fake 17-year-old boy's account was recommended a video of Andrew Tate. The content featuring controversial misogynistic influencer Andrew Tate grew in frequency over the duration of experiment. Of the last 30 pieces of content that Reset.tech examined, 29 featured Andrew Tate," she told parliament.

"This experiment took under two hours to complete, meaning that a teenage boy showing an interest in the work of Jordan Peterson would encounter Andrew Tate within about an hour. If that teenage boy then watches or likes the Andrew Tate content by the two-hour mark, the system will almost exclusively feed him Tate content. So we can see that as our sons or daughters sit on the couch, heads down on social media, they're anything but safe."

It's time to end the moral panic and engage in reality — young people don't live in a binary where the online world is separate to the real world.

There is a political consensus across the parliament that there is an urgency for reform that actually works and holds big tech to account.

Patricia Karvelas is the presenter of RN Breakfast and co-host of the Party Room podcast. She also hosts Q+A on ABC TV Mondays at 9.35pm.

Enhance your entire business with deeper social insights

The value of social data extends much farther than social media— if it’s accessible. Without that critical knowledge, you’re left playing catch up. Sprout combines social listening and performance data to help your team make an impact and prove it.

#1 Best Software, 2024

Zero in on the metrics you need to impact every part of your organization

Learn from pivotal customer conversations

Quickly ingest high volumes of business-critical insights to help enhance your ability to pivot.

Identify new opportunities

Surface notable trends, find gaps in your industry, track consumer behavior and understand buyers’ attitude toward competitors to uncover new business opportunities.

Understand customers better

Stay ahead of a crisis

Prove the ROI of social media

Connect the dots between social initiatives and larger company goals so you can prove the true impact of your social efforts with Premium Analytics.

Tailor metrics to align with business goals

Go beyond surface-level numbers and define metrics based on the priorities of unique, individual teams in your organization.

Tell a complete story with your data

Share your performance with stakeholders

Salesforce Marketing Cloud Intelligence

Give data-driven marketers their dream. Incorporate owned performance across all your social networks into overall marketing dashboards, enriched with Sprout tag and campaign data.

Make the business right decisions by combining Sprout’s robust social data with other key business channels for a comprehensive view of your data in Tableau.

“ The team at Sprout helped us realize we could maximize even more potential, walking us through what data might be useful based on their expertise. It’s not just the data, but the knowledge and experience and best practices Sprout brings, not only from an industry perspective, but from outside it as well.”

Simon Kopec, Director of Social Media & Content Strategy at Loew’s Hotels

Accelerate your brand growth with complete social data

Raygun becomes viral sensation during breaking performance at 2024 Paris Olympics: Social media reacts

Breaking , more commonly known as breakdancing, made its debut as an Olympic sport this week at the 2024 Paris Games , with 17 B-girls and 16 B-boys making their way to France with the hopes of securing a gold medal.

On the first day of competition, viewers from across the world were treated to a different kind of introduction — not to the sport itself, but one of its athletes.

Though she was a long way from winning a gold medal, likely no breaker Friday captured the imagination of the international audience more than Rachael Gunn, an Australian breaker who competes under the name “Raygun.”

REQUIRED READING: Follow USA TODAY's coverage of the 2024 Paris Olympics

Raygun went 0-3 in her head-to-head competitions Friday — falling to Logistx of the United States, Syssy of France and eventual silver medalist Nicka of Lithuania by a combined score of 54-0 — and failed to record a point across those three matches, but for what she lacked in smoothly executed moves, she made up for in the hearts she won over with her demeanor.

Raygun’s short-lived Olympic experience made her a celebrity, one who people became even more enamored with once they learned more about her.

The 36-year-old Gunn, who was one of the oldest qualifiers in the breaking competition, has a PhD in cultural studies and is a college professor at Macquarie University in Sydney. Her research focuses primarily on breaking, street dance and hip-hop culture while her work draws on “cultural theory, dance studies, popular music studies, media, and ethnography.”

“In 2023, many of my students didn’t believe me when I told them I was training to qualify for the Olympics, and were shocked when they checked Google and saw that I qualified,” Gunn said to CNBC earlier this month .

Unlike much of her competition in Paris, Gunn took up break dancing later in life. She didn’t enter her first battle until 2012.

On Friday, a person who began the day as a little-known academic ended it as a viral worldwide sensation.

Here’s a sampling of the reaction to Raygun and her performance:

2024 PARIS OLYMPICS: Meet the members of Team USA competing at the 2024 Paris Olympics

Social media reacts to Raygun’s breaking performance at 2024 Paris Olympics

I could live all my life and never come up with anything as funny as Raygun, the 36-year-old Australian Olympic breakdancer pic.twitter.com/1uPYBxIlh8 — mariah (@mariahkreutter) August 9, 2024

Give Raygun the gold right now #breakdancing pic.twitter.com/bMtAWEh3xo — n★ (@nichstarr) August 9, 2024

my five year old niece after she says “watch this!” : pic.twitter.com/KBAMSkgltj — alex (@alex_abads) August 9, 2024

I'd like to personally thank Raygun for making millions of people worldwide think "huh, maybe I can make the Olympics too" pic.twitter.com/p5QlUbkL2w — Bradford Pearson (@BradfordPearson) August 9, 2024

The Aussie B-Girl Raygun dressed as a school PE teach complete with cap while everyone else is dressed in funky breaking outfits has sent me. It looks like she’s giving her detention for inappropriate dress at school 🤣 #Olympics pic.twitter.com/lWVU3myu6C — Georgie Heath🎙️ (@GeorgieHeath27) August 9, 2024

There has not been an Olympic performance this dominant since Usain Bolt’s 100m sprint at Beijing in 2008. Honestly, the moment Raygun broke out her Kangaroo move this competition was over! Give her the #breakdancing gold 🥇 pic.twitter.com/6q8qAft1BX — Trapper Haskins (@TrapperHaskins) August 9, 2024

my dog on the lawn 30 seconds after i've finished bathing him pic.twitter.com/A5aqxIbV3H — David Mack (@davidmackau) August 9, 2024

My wife at 3AM: I think I heard one of the kids Me: No way, they are asleep *looks at baby monitor* pic.twitter.com/Ubhi6kY4w4 — Wes Blankenship (@Wes_nship) August 9, 2024

me tryna get the duvet off when i’m too hot at night #olympics pic.twitter.com/NM4Fb2MEmX — robyn (@robynjournalist) August 9, 2024

Raygun really hit them with the "Tyrannosaurus." pic.twitter.com/ZGCMjhzth9 — Mike Beauvais (@MikeBeauvais) August 9, 2024

Raygun (AUS) https://t.co/w2lxLRaW2x — Peter Nygaard (@RetepAdam) August 9, 2024

The Australian Olympian 'Raygun' went viral for her breaking moves. Now she's defending them.

• Rachael Gunn, known as "Raygun," is an Australian B-girl (break-girl) who competed at the Olympics .
• She lost three battles in the round-robin part of the competition, but her moves went viral online.
• Gunn and sporting organizations are speaking out about harassment and misinformation after her performance.

Breaking made its debut at the 2024 Paris Olympics — and while she didn't earn a spot on the podium, the Australian breaker Rachael Gunn, known as Raygun, has received plenty of recognition online.

Gunn is a 36-year-old lecturer at Macquarie University in Sydney whose research focuses on the "cultural politics of breaking," according to her faculty profile .

But Gunn's time on the Olympic stage was short-lived. The B-girl was eliminated during the round-robin stage of the women's breaking competition, losing in one-on-one battles to the United States' Logistx, France's Syssy, and Lithuania's Nicka.

Raygun didn't earn a point in any of those battles, but as clips of her performance spread online, she got something else: instant meme status.

Here's what you need to know about Raygun now that the breaking competition is over.

Raygun is an academic who studies breaking — and she competes internationally

Before Gunn went to the Olympics, she approached the 2024 Games from an academic perspective.

With her coauthor, Lucas Marie, Gunn published an article in the June 2023 issue of Global Hip Hop Studies titled "The Australian breaking scene and the Olympic Games: The possibilities and politics of sportification." The article examined how the Olympics' institutionalization would affect the Australian breaking scene.

Alongside her academic career, Gunn is a competing B-girl. But before she got into breaking, she had experience with ballroom dancing, jazz, hip-hop, salsa, and tap, The Australian Women's Weekly reported. Gunn told The Sydney Morning Herald that her husband, Samuel Free, introduced her to breaking in 2008 while they were at university. Free is still her coach, she said.

Gunn told Women's Weekly that breaking "hooked" her in 2012, around the time that she began her doctoral program in cultural studies. She began competing more seriously in 2018 and eventually set her sights on the Olympics.

According to her university profile, she was the top-ranked B-girl of the Australian Breaking Association in 2020 and 2021, representing the country at the World DanceSport Federation Breaking Championships in 2021, 2022, and 2023. She also won the WDSF Oceania Breaking Championships in 2023.

"My bag always has two main things: It's like, my knee pads and my laptop," Gunn said on the podcast " The Female Athlete Project ." "Because I need my knee-pads to break. And then, yeah, just do some emails quickly. Or like, do some revisions on a chapter I submitted, or copyedit this article I did, or moderate those grades."

The athlete also told the Herald that she preferred to wear "baggy jeans and a baggy T-shirt" while breaking.

"I like the heaviness they bring," Gunn said. "Maybe it's my background in hip-hop, but having weight closer to the ground works for me, gets me in the right headspace."

Raygun's performances at the Olympics sparked memes and criticism

Raygun took the stage at the Olympics wearing a tracksuit in Australia's green and gold, breaking out moves that included hopping like a kangaroo. Her performances attracted attention online and memes that compared her moves to, among other things, dancing children.

Related stories

The fact that RayGun has a Ph.D in breakdancing is its own commentary on academia vs real world expertise. https://t.co/pQcL8HzAW9 — BioTechSnack (@SnackBioTech) August 9, 2024

me forcing my mom to watch the dance i made up in the pool pic.twitter.com/zbtwEFjpTG — kenzi (@kenzianidiot) August 9, 2024

Judges made the right call here because what was that move lol #Olympics #Breakdancing pic.twitter.com/sXAs9AdHjX — MⓞNK BLOODY P👑s (@MonkeyBlood) August 9, 2024

But some critics argued that Raygun's performance didn't represent breaking — a sport that will not return to the 2028 Games in Los Angeles.

Breaking came from Black and brown communities in the Bronx in the 1970s. Malik Dixon, an African American man who lives in Australia, told the Australian Broadcasting Corporation that Gunn came off as "somebody who was toying with the culture" during a significant moment for the sport.

(You can watch the 2024 Olympic events — including Raygun's full performance — on Peacock.)

Raygun qualified for the Olympics through the Oceania Breaking Championships

There were three ways to qualify for breaking at the Olympics, which the World Dance Sport Federation (WDSF) outlined in April 2022: at the WDSF championship in Belgium in September 2023, in a continental qualifier, or in an Olympic qualifier series held in 2024. Gunn qualified regionally by winning the WDSF Oceania Breaking Championships, which were held in Sydney in October 2023.

AUSBreaking organized the Oceania Breaking Championships, according to the WDSF .

AUSBreaking posted on Instagram about the Oceanic Olympic qualifying event on Instagram in September 2023, announcing in a September 25, 2023 post that competitor registration was open. The panel of judges was composed of 10 breakers from multiple countries, led by head judge Katsu One of Japan.

Per the Sydney Morning Herald, Gunn was the highest-scoring B-girl on day one of the championships. She won two battles on the second day to secure her title and a qualifying spot in the 2024 Olympic Games in Paris.

AUSBreaking released a statement on Instagram Monday about the selection process, saying that the qualifying event was "open to all interested participants in the Oceanic region," conducted in line with WDSF standards, and adjudicated by an international panel that used the same judging system as the 2024 Olympics.

"Ultimately, Rachael Gunn and Jeff Dunne emerged as the top performers in exactly the same process, securing their spots to represent Australia in Paris," the statement reads. "Their selection was based solely on their performance in their battles on that day."

Raygun and sporting organizations have spoken out about misinformation after her performance

Claims have circulated online that Gunn unfairly obtained her spot in the games. Posts online, as reported by the Australian Associated Press , claimed that Gunn's husband was one of the judges in her qualifying event. One petition hosted on Change.org claimed that she established the governing body that ran the selection process. That petition was eventually removed after it was placed under review, per an archived snapshot .

A representative for Change.org confirmed to Business Insider on Thursday that the petition had been flagged for misinformation, reviewed per the platform's community guidelines, and removed from the platform.

"Change.org maintains strict guidelines against content that constitutes harassment, bullying, or spreading false information. We take such matters seriously and remove any content that violates these standards to protect our users and uphold the integrity of our community," the rep said in an email statement to BI.

Despite the online claims, Free was not one of the judges at Gunn's qualifying event. And Gunn did not establish AUSBreaking. The organization said in a statement that it was founded by its president Lowe Napalan in 2019, and "at no point" was Gunn "the founder, an executive, committee member, or in any position of leadership."

The Australian Olympic Commission (AOC) also released a statement condemning the Change.org petition, and demanding its removal. It also said that by winning the Oceania championship, Gunn was "legitimately nominated" by DanceSport Australia to the AOC to represent Australia at the Olympics.

"The petition has stirred up public hatred without any factual basis. It's appalling," AOC chief executive officer Matt Carroll said in the statement. "No athlete who has represented their country at the Olympic Games should be treated in this way and we are supporting Dr. Gunn and Anna Meares at this time."

In a video uploaded to her personal Instagram account, Gunn said that she was "honored" to have represented Australia and breaking during its Olympic debut. But the "hate" that followed was "devastating," she said. When it came to misinformation around her qualification, Gunn referred viewers to previously issued statements from the AOC and AUSBreaking.

Raygun and breaking judges have defended her Olympic performance

At a press conference on Saturday, the day after Gunn's competition, Anna Meares, the head of the Australian team, responded to criticism of Gunn online.

"I love Rachael, and I think that what has occurred on social media with trolls and keyboard warriors, and taking those comments and giving them airtime, has been really disappointing," Meares said, per ESPN .

"Raygun is an absolutely loved member of this Olympic team. She has represented the Olympic team, the Olympic spirit with great enthusiasm. And I absolutely love her courage," Meares continued. "I love her character, and I feel very disappointed for her, that she has come under the attack that she has."

During a press conference on Sunday, Martin Gilian, the Olympic breaking head judge, defended Gunn's performance, saying breaking was "all about originality" and representing your roots, the Australian Broadcasting Corporation reported.

"This is exactly what Raygun was doing," Gilian said. "She got inspired by her surroundings, which in this case, for example, was a kangaroo."

Gunn said during the Saturday press conference that "all of my moves are original," ESPN reported. She told The Guardian that her biggest strength was "creativity."

"I was never going to beat these girls on what they do best, the dynamic and the power moves, so I wanted to move differently, be artistic and creative," Gunn told The Guardian, "because how many chances do you get that in a lifetime to do that on an international stage. I was always the underdog and wanted to make my mark in a different way."

This story was originally published on August 12, 2024, and has been updated to include the latest information and statements from those involved.

• Main content