The Attention Economy

A human gets about 16 waking hours per day, and nearly every internet business is bidding for slices of that fixed inventory. Herbert Simon saw the trade in 1971: abundant information creates scarce attention. After 2010, feeds, notifications, recommendation systems, and ad auctions turned that scarcity into an industrial market.

The product is not the app. The product is the predictable redirection of attention.

The mechanism

The attention economy has three moving parts.

First, attention is finite. A platform cannot create a 25th hour in the day; it can only take minutes from sleep, work, family, boredom, books, or rival apps. The average American adult spends roughly 2 hours per day on social media, depending on survey and year. Over 50 years, 2 hours daily becomes more than 4 waking years.

Second, attention can be priced. Google search ads, Meta feed ads, YouTube pre-rolls, TikTok shops, newsletter sponsorships, and creator subscriptions all convert attention into money. The unit changes: click, impression, watch time, retention, paid conversion. The business question stays the same: how much future behavior can this moment of attention predict?

Third, attention can be tuned. A feed can test 10 thumbnails, 20 ranking weights, or 1,000 notification variants faster than a newspaper editor could test one front page. The machine does not need a theory of human nature. It needs a metric and enough traffic.

The engineering stack

The post-2010 feed has a small set of repeatable parts:

Design choice	What it changes
Infinite scroll	Removes the natural stopping point
Variable rewards	Makes the next refresh uncertain enough to repeat
Public counters	Turns reading into status comparison
Push notifications	Reopens the app from outside the app
Recommendation ranking	Replaces chronology with predicted engagement
Short video	Lowers the cost of one more unit of attention

None of these parts requires evil intent. A product team can chase session length, retention, and ad yield while telling itself it is only giving users what they want. The harder question is whether revealed preference inside a persuasion machine still counts as preference.

What's contested

The strongest claims about harm are not all equally settled. Heavy social media use correlates with worse adolescent mental health in many datasets, with stronger concern around teenage girls after about 2012. The causal story is harder: depressed teenagers may use social media more, social media may worsen depression, or both may be true in different groups.

The same problem appears in attention research. People report shorter focus and more distraction, but measurement is messy. A 45-second clip habit may train shallower attention, or it may mostly attract people already avoiding longer tasks. The honest position is not “phones ruined minds”; it is that billions of daily experiments are being run on attention faster than public evidence can settle their effects.

Why this has to do with other realms

The attention economy is ad-tech, but it is also applied biology. Variable reward schedules come from behaviorist psychology, dopamine is often invoked too loosely, and sleep loss turns a media question into a body question. A feed competes not just with another feed, but with circadian rhythm, hunger, solitude, and concept flow state.

It also belongs beside money and power. Creator income follows concept power laws: a tiny fraction of accounts capture most of the audience, while millions produce for almost no pay. That makes the attention economy a cousin of markets, status games, and concept second brain as a private defense against public distraction.

Counter-moves

The cleanest personal counter-move is architecture, not willpower: remove default notifications, keep the phone outside the bedroom, use scheduled access, make long reading easier than short checking. These changes work because they alter the choice environment before desire arrives.

The institutional counter-moves are slower. The EU Digital Services Act began applying major platform duties in 2024, including researcher access and transparency rules for large online platforms. Subscription products such as Spotify Premium, YouTube Premium, and paid newsletters shift part of the incentive from ad exposure to retention, but retention can still reward compulsion.

Open question

If an AI assistant can filter, summarize, rank, and answer before a user enters the feed, does it weaken the attention economy, or does it become the next bidder for the same 16 waking hours?

Key Sources

Herbert A. Simon, “Designing Organizations for an Information-Rich World” (1971) — the canonical line on information abundance and attention scarcity.
B. F. Skinner, Science and Human Behavior (1953) — the behaviorist foundation behind reinforcement schedules.
Thomas H. Davenport and John C. Beck, The Attention Economy (2001) — early business framing of attention as a scarce asset.
Tim Wu, The Attention Merchants (2016) — history of advertising-funded attention capture from newspapers to the web.
Allcott et al., “The Welfare Effects of Social Media” (American Economic Review, 2020) — randomized evidence from Facebook deactivation.
Orben and Przybylski, “The association between adolescent well-being and digital technology use” (Nature Human Behaviour, 2019) — careful treatment of effect sizes and interpretation.

Abhishek's take

I see this on the floor when a line sheet loses to a notification before the buyer has finished the first page. The tools I wrote do not try to make people look busy; they pull one vendor exception to the top, then get out of the way. A buying floor runs on attention as much as budget.

Abhishek