Case Study 10.1: The Like Button — From "Awesome Button" to Global Approval Machine

A Simple Idea in a Complex System

In early 2007, Justin Rosenstein was 24 years old and working as a product engineer on Facebook's platform team. He had an idea that seemed simple, even obvious: what if users could express appreciation for a post with a single click, without having to write a comment? Something that said "I see this, and I like it," frictionlessly, publicly, immediately.

He called it the "awesome button." The concept sketched itself in a few lines: a small icon below each post, a click that registered a visible count of positive responses, a notification to the original poster that someone had responded positively. No text required. No commitment beyond a single tap.

Fifteen years later, that tap has been made approximately ten trillion times, across Facebook, Instagram, Twitter, TikTok, LinkedIn, and dozens of other platforms that adopted variants of the mechanism Rosenstein proposed. The like button — or its platform-specific variants: the heart, the star, the thumbs-up — is the most universal action on the social internet. It is the grammar of digital social feedback.

Understanding how a 24-year-old's simple idea became one of the most psychologically consequential design decisions in the history of technology requires understanding both the institutional dynamics at Facebook in 2007 and the behavioral mechanisms the button was about to activate at global scale.

The Two-Year Debate

Rosenstein's "awesome button" proposal generated significant internal discussion at Facebook. According to later accounts from multiple people who were present, the debate was genuinely substantive — not a brief discussion quickly dismissed or quickly approved, but an ongoing consideration that occupied the product team for approximately two years before the button launched in February 2009.

The concerns raised internally were prescient. Would a like button make users feel judged by the count on their posts? Would it incentivize people to post for approval rather than connection? Would the visibility of like counts create a social hierarchy where users with more followers systematically accumulated more likes, creating a feedback loop that entrenched popularity? Would people who received low likes feel stigmatized?

Against these concerns, proponents argued that the like button was additive: it gave users a new way to express positive feelings with minimal effort, which would increase the total volume of positive feedback exchanged on the platform. Users with low-engagement posts would see some validation that previously required a comment to express. Users with high-engagement posts would see their content affirmed. The net social effect would be more positive.

The resolution of these debates — in favor of shipping the like button — reflects the institutional dynamics of product development at a growth-focused technology company in 2009. The metric that mattered most to Facebook at the time was engagement: daily active users, time spent on site, content interactions. The like button, predictably, increased all of these metrics dramatically once it shipped. Concerns about psychological effects were real but abstract; the engagement data was concrete and immediately visible.

According to accounts from people familiar with the internal decision-making, including the reporting of journalist Paul Lewis in The Guardian and subsequent accounts in various books about Facebook's development, the decision to add the like button was made without detailed empirical research on its psychological effects. The harm concerns were speculative; the engagement benefits were predictable. In the institutional calculus of a fast-growing tech company, the speculative concerns lost.

Launch and the Engagement Explosion

The like button launched on February 9, 2009, alongside a redesigned Facebook interface. The response was immediate and dramatic. Within days, the number of explicit interactions on the platform — clicks, comments, reactions — increased sharply. Posts that had previously attracted only comments began attracting likes, and posts that had attracted no response at all now attracted at least a few. The feedback density of the platform increased.

For Facebook's product team, this was a clear win. More engagement meant more time on site, more advertising impressions, more user investment in the platform. The like button was quickly recognized internally as one of the most impactful product changes in Facebook's history to that point.

For users, the change was subtler but equally significant. The introduction of the like count created a new social metric — a visible number attached to every post, publicly accessible, updating in real-time. Users began, predictably, paying attention to these numbers. Posting decisions began to reflect consideration of likely like counts. The social dynamics of the platform shifted: it was no longer simply a place to share things with friends, but a place where what you shared was subject to a public numerical verdict.

The behavioral patterns that researchers would spend the next decade studying — compulsive like-checking, posting decisions shaped by anticipated approval, emotional responses calibrated to like counts, comparison anxiety driven by visible social metrics — began developing within weeks of the button's introduction. They were not invisible at the time: anecdotal reports of users refreshing pages to watch like counts accumulate, of users feeling anxious about low-engagement posts, of users deleting posts that hadn't performed well enough, circulated almost immediately. But these were treated as curiosities rather than warning signs.

Instagram, Twitter, and the Spread

The like button concept spread rapidly after Facebook's success with it. Instagram, launched by Kevin Systrom and Mike Krieger in October 2010, included a heart-based appreciation mechanism from its first day — the "double-tap to like" gesture became one of the platform's signature interactions, something so embedded in the experience that it defined how users related to content on the platform.

When Facebook acquired Instagram for approximately $1 billion in April 2012, the two like systems merged conceptually though not technically. Instagram's visually focused, like-count-prominent design gave the quantified approval mechanism a new and particularly powerful context: beautiful images presented alongside their numerical approval verdicts, creating what would become the most potent comparison machine in social media history.

Twitter's path was different. The platform had a "favorite" feature (star icon) from early in its history, and this remained relatively stable until November 2015, when CEO Jack Dorsey announced the conversion of favorites to likes, with the star icon replaced by a heart. The change was motivated partly by international user testing — the heart, Dorsey explained, was more universal and intuitive across cultures than the concept of "favoriting." But internal data on user engagement with the heart versus star format, which reportedly showed significantly higher interaction rates with heart icons, was also a factor.

TikTok, when it launched its global version in 2018, adopted the heart-based like system essentially unchanged from the Instagram model. The heart — a small icon below each video, tapped to express appreciation, displaying a count of all taps — is the primary feedback signal on a platform that by 2023 had over 1.5 billion monthly active users. The "awesome button" concept, in various heart-shaped forms, is now the primary feedback language of the entire short-form video medium.

The Regret of the Builders

The most striking dimension of the like button's story is the candor with which many of its creators and early developers have discussed their concerns about what they built.

Justin Rosenstein, the original "awesome button" proposer, gave a detailed account of his views in Paul Lewis's September 2017 Guardian profile, "Our minds can be hijacked: the tech insiders who fear a smartphone dystopia." He described the like button as intended to "spread positivity" but acknowledged the evidence that it had created dynamics he found troubling. He now uses software to limit his own social media use and has spoken at conferences about the behavioral effects of the systems he helped build. He has been careful not to claim sole responsibility for the like button's effects — the entire design and business ecosystem around engagement optimization contributed — but he has not avoided the subject.

Leah Pearlman, who as a product manager at Facebook was closely involved in the like button's implementation and public launch, has spoken even more personally. In a 2017 interview with The Guardian, she described developing a compulsive relationship with the like button she had helped create: checking her posts repeatedly for new likes, experiencing genuine anxiety when engagement was lower than expected, and hiring a coach to help her manage the emotional dynamics that her social media use had created. "I was at a point where I needed external validation to feel good about myself — with every post I was making," she said.

The experience was, she noted, personally instructive: "When I noticed that happening, I also noticed: I was addicted to this thing, and I was the one who made it." The recognition that she had been caught in the mechanism she had helped design — that the system she built had been calibrated to produce exactly the behavior she found herself engaging in — is a remarkable statement from someone with her proximity to the decision.

Pearlman subsequently left Facebook and has written and spoken about digital wellness, specifically about the relationship between social media design and human psychological health. She has not simply criticized her former employer; she has reflected on the systemic nature of the problem — that the like button was one product decision in a commercial system with powerful incentives that consistently rewarded engagement maximization over user wellbeing.

The "regret" narratives of like button creators are worth examining carefully. They do not constitute proof of harm — individuals can be wrong about the effects of the systems they built, just as they can be wrong in other directions. But they are evidence that people with the deepest knowledge of how the like button was designed and what it was designed to do have concluded, with the benefit of observation and reflection, that the effects include dynamics that were harmful and that were at least partially foreseeable.

The Research on Effects: What We Know

The research on the like button's effects is now substantial. Several findings are well-established across multiple methodologies.

Quantified social feedback changes posting behavior. Studies using controlled experiments have found that users given access to like counts on their posts behave differently than users without this information: they post more frequently, optimize their content toward engagement, and show more sensitivity to under-performing posts. The quantification creates a feedback loop between approval signals and content production that does not exist without the visible count.

Visible like counts amplify social comparison. Research consistently finds that environments with visible social approval metrics produce more frequent and more intensive social comparison than environments without them. Upward social comparison — comparing yourself to those with higher engagement — predominates and is associated with decreased wellbeing.

The adolescent effects are stronger. The research on Instagram's effects, particularly on teenage girls, consistently finds larger negative associations with wellbeing outcomes for adolescents than for adults. This is consistent with the developmental amplification of social reward sensitivity described in Chapter 10 and with the internal Meta research that found significant body image effects specifically in teenage girls.

Hiding counts does not fully resolve the dynamics. Instagram's 2019 experiment with hidden like counts showed that while explicit like-count comparison decreased, users found proxy metrics (follower counts, comment counts, story view counts) to fulfill the comparison function. The underlying comparison drive, trained over years of quantified social feedback, does not disappear when one number is hidden.

Experimental reductions in social media use improve wellbeing. Multiple studies using experimental designs — randomly assigning participants to reduce social media use versus continue as usual — have found wellbeing improvements in the reduced-use conditions. This causal evidence is the strongest available for the conclusion that the social reward environment of platforms including the like button contributes to wellbeing costs.

What the Like Button Built

Looking at the fifteen years since February 9, 2009, the like button's legacy is not simply the engagement metric it created. It is a behavioral environment that billions of people now inhabit, in which:

• Social approval is expressed as a number
• That number is public and persistent
• The number updates in real-time, creating an ongoing monitoring incentive
• Comparison is facilitated by the numbers' universality and visibility
• Content decisions are shaped by approval prediction
• Emotional wellbeing is partly mediated by approval receipts

Justin Rosenstein wanted to make it easier to express appreciation. He did. He also, with the institutional support of one of the most powerful technology companies in history, built the infrastructure for a global approval economy that has restructured how hundreds of millions of people — a significant portion of whom are adolescents — experience social validation.

The button is small. The button is everywhere. And the button, as its own builders have acknowledged, does something more complex than it looks.

Sources: Lewis, P. (September 6, 2017). "Our minds can be hijacked: the tech insiders who fear a smartphone dystopia." The Guardian. Additional context from Silverman, J. (2015). Terms of Service: Social Media and the Price of Constant Connection. Harper Perennial; and Lanier, J. (2018). Ten Arguments for Deleting Your Social Media Accounts Right Now. Henry Holt and Company. On the research on like button effects, see Valkenburg, P.M., Fardouly, J., & Nesi, J. (2022). Social media use and its impact on adolescent mental health: An umbrella review of the evidence. Current Opinion in Psychology.