The Like Button Turned Social Approval Into a Scoreboard

Post something online.

Then leave.

A few minutes later, come back.

12 likes.

Refresh.

19.

Later, someone else posts something similar.

437.

The number cannot tell you exactly what anyone thought.

Maybe they loved it. Maybe they barely noticed it. Maybe they clicked because everyone else seemed to.

But the number still does something remarkable.

It takes an invisible social reaction and gives it a score.

Humans did not need the internet to invent approval.

We have always looked for signs that other people notice us, accept us, enjoy what we make, or agree with what we say.

The internet changed something else.

It changed the interface of approval.

Most social feedback used to be difficult to measure.

A room laughed.

Someone smiled.

A friend invited you back.

A conversation lasted longer than expected.

People talked about your work.

You developed a reputation.

These signals could matter enormously, but they were scattered across people, places, and moments.

There was no universal counter beneath the interaction.

Online platforms changed that.

A social response that once existed mostly as a feeling could now leave behind a visible unit.

Facebook launched the Like button in February 2009.

It was a tiny interface element.

But it helped normalize a very different way of reading social response:

as something that could be counted.

At first, a like count seems passive.

People saw something.

Some responded.

The number records what happened.

But the number is also visible to the next person who arrives.

That changes its role.

In a 2016 experiment using an Instagram-like feed, adolescents were shown photographs with either many or few likes.

The photographs stayed the same.

The popularity signal changed.

Participants were more likely to like photographs when they believed many other people had already liked them. The study also found different neural responses when participants viewed highly liked images, particularly in regions associated with reward processing, social cognition, imitation, and attention.

The count was doing two things at once.

It recorded previous approval.

And it became information for the next observer.

Once visible, the metric is no longer merely describing the social environment.

It has entered it.

Quantification creates another possibility:

comparison.

Your post against yesterday’s.

Your photo against someone else’s.

One creator against another.

A person does not need an analytics dashboard to understand:

84 < 840

Social approval had always been uneven.

The difference now was that some of it could be expressed in a shared unit.

That makes vague social differences easier to see.

And easier to compare.

A laugh cannot be cleanly ranked against another laugh.

A compliment does not come with a public total.

A like does.

The scoreboard does not simply tell you that people responded.

It tells you how much response accumulated—and places that number beside every other score.

Then comes the deeper change.

If a metric repeatedly follows behavior, people can begin learning from the metric.

A 2021 study analyzed more than one million posts from 4,168 users across four social platforms, including Instagram. The researchers found that posting behavior followed patterns consistent with reward learning: higher rates of social reward were associated with shorter intervals before people posted again.

But observational data cannot establish the entire causal story.

So the researchers recreated part of the system in a controlled experiment.

176 participants could post memes and receive likes from what they believed were other users. The researchers changed the average amount of social reward participants received.

When the reward rate was higher, participants posted again sooner.

The difference in posting latency between the high- and low-reward conditions was about 10.9%.

That does not prove people redesign every photograph, joke, or opinion to maximize likes.

The study did not test that.

What it shows is narrower:

the metric can affect what happens next.

The scoreboard can talk back to the player.

For professional creators, measurements become difficult to ignore.

They are not merely signals of audience reaction. They can become part of how creators understand visibility, commercial value, and what to do next.

A study based on interviews with 35 fashion and lifestyle creators found that some used Instagram statistics to change publishing strategies—for example, choosing upload times to increase engagement, likes, or comments. The same research describes creators adapting their production practices as platform features and visibility systems changed.

This is qualitative evidence, not proof that a particular number directly caused a particular creative decision.

But it reveals a different working environment.

The audience responds.

The platform converts some of those responses into numbers.

The creator sees the numbers.

The next decision is made with those numbers available.

Creation now has an instrument panel.

And once performance can be measured immediately, ignoring the measurement becomes a choice too.

Eventually, the metric itself became a design problem.

Instagram experimented with hiding public like counts and, in 2021, Meta gave users the option to hide public counts on both Instagram and Facebook.

Meta said it had tested the change to see whether hiding counts could make the Instagram experience feel less pressured. But the company also reported mixed reactions: some people found hidden counts beneficial, while others disliked losing a signal they used to judge what was popular or trending.

That tension reveals what the number had become.

It could create pressure.

But it also carried information.

The same number could function as feedback, social proof, comparison, and navigation.

Remove the scoreboard and you remove some of all four.

Independent research complicates the easy story that hiding likes automatically solves the problem.

An experiment involving 280 Instagram users manipulated both the amount of likes participants received and whether those likes were visible to other people. The effects on negative affect, loneliness, and related outcomes depended on the combination of like quantity and visibility rather than following a simple rule in which public counts were always harmful and hidden counts were always beneficial.

So this is not an argument that likes are inherently bad.

Nor is the hidden system simply:

numbers make people unhappy.

The deeper transformation happened earlier.

Social approval became measurable infrastructure.

Once that happened, a number could operate simultaneously as:

feedback,

social information,

reward,

comparison,

and an input into future behavior.

The desire for approval is older than technology.

So are status, imitation, comparison, and social influence.

The like button created none of them.

What platforms did was compress part of that messy social world into a standardized signal.

A click.

A heart.

A number.

And numbers behave differently from feelings.

They can be ranked.

Compared.

Aggregated.

Displayed.

Analyzed.

Used to evaluate performance.

Fed back into future decisions.

Watched rise in real time.

A room full of people might once have left you with the vague impression that something went well.

The internet can tell you:

2,413 people clicked Like.

The number is more precise.

Its meaning is not.

Someone may genuinely love the post.

Someone may be acknowledging a friend.

Someone may be following a crowd.

The interface compresses those different motives into the same visible unit.

And that is precisely what makes the number powerful.

Once a messy social response becomes standardized, people can begin responding not only to one another—

but to the measurement itself.

The like button did not invent our need for approval.

It gave approval a scoreboard.

And once the scoreboard became visible, it stopped merely recording the game.

It became part of how the game was played.

Facebook — One Billion: Key Metrics — Facebook’s official historical metrics, including the launch of Likes in February 2009.

The Power of the Like in Adolescence — Sherman et al., Psychological Science. An experiment examining how visible peer endorsement influenced responses to social-media content.

A Computational Reward Learning Account of Social Media Engagement — Lindström et al., Nature Communications. Research on more than one million posts, followed by a controlled experiment on social reward and posting behavior.

“You Need At Least One Picture Daily, if Not, You’re Dead”: Content Creators and Platform Evolution in the Social Media Ecology — Arriagada & Ibáñez. Research based on interviews with 35 content creators about changing platform practices, metrics, and visibility.

Giving People More Control on Instagram and Facebook — Meta’s explanation of its experiments with hiding public like counts and the controls introduced in 2021.

Hiding Instagram Likes: Effects on Negative Affect and Loneliness — Wallace & Buil, Personality and Individual Differences. An experiment examining how like quantity and visibility affected users’ experiences.