Impactism
日本語 Join Impactism
Manifesto The Impact Standard Foundations Objections Read For organizations 日本語
Join Impactism
A man's face lit from below by the screen of the phone he is holding in the dark
Photo by Adrian Swancar on Unsplash

Essay

What the Attention Economy Actually Is

October 1, 20268 min read

The attention economy is the arrangement in which human attention, rather than information, is the scarce good that gets competed for and sold. Information can be copied at no cost, so there is effectively an unlimited supply of it. The hours a person has available to notice anything are fixed and cannot be manufactured. Once that is true, the thing worth fighting over stops being what you can say and becomes whether anyone will listen.

Herbert Simon set out the mechanism in 1971, before the web existed. Michael Goldhaber named the economy in 1997 and made a prediction about it that turned out to be half right in an instructive way. This essay walks the argument from those two, through what the business actually became, to what a scoreboard built on noticeability selects for, and why it takes the seat whenever nothing else is installed there.

The mechanism, written down in 1971

Simon’s passage is usually quoted in one clipped line. The whole thing is better, because the last clause is the part that matters:

“In an information-rich world, the wealth of information means a dearth of something else: a scarcity of whatever it is that information consumes. What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it” (Simon 1971, in Greenberger, ed., Computers, Communications, and the Public Interest, Johns Hopkins Press).

Simon was writing about organisational design, not media. He did not use the phrase “attention economy”. What he supplied was the premise every later argument rests on: attention is a budget, and a budget has to be allocated. He did not say who would do the allocating.

The name, and the prediction attached to it

Michael Goldhaber supplied both, in First Monday in April 1997. His starting move is the same one Simon made and he is blunt about it: “economies are governed by what is scarce, and information, especially on the Net, is not only abundant, but overflowing.” From which: “Attention, at least the kind we care about, is an intrinsically scarce resource” (Goldhaber 1997).

One clarification, because the two terms get used as synonyms and are close to opposites. An “information economy” would be an economy whose scarce good is information. Goldhaber’s answer to that is flat: information on the net “is not in any way scarce, and therefore it is not an information economy towards which we are moving.” What is scarce is the noticing. Calling the same thing an information economy points every strategy in the wrong direction, at producing more, when the binding constraint is on the receiving end.

He then went further than most people remember. He argued the attention economy would come with “its own kind of wealth, its own class divisions - stars vs. fans - and its own forms of property, all of which make it incompatible with the industrial-money-market based economy it bids fair to replace.” He thought attention would displace money.

Half of that was right and the wrong half is the useful one. The stars-and-fans class structure arrived exactly as described. The replacement did not. Attention did not become an alternative to the money economy. It became an input to it.

What the business actually turned out to be

Tim Wu wrote the history of that input. The Attention Merchants traces the model back to 1833, when, as the publisher’s description puts it, “Benjamin Day discovered he could get rich selling newspapers for a penny.” Day was not selling newspapers to readers at a profit. He was selling readers to advertisers. The formula Wu describes is “free diversion in exchange for a moment of your time, sold in turn to the highest-bidding advertiser,” and his argument is that “every new medium - from radio to television to Internet companies such as Google and Facebook - has attained commercial viability and immense riches by turning itself into an advertising platform” (Wu 2016).

So the honest description of the attention economy is not a successor to the money economy. It is a supply chain. Attention is harvested at one end, in units of a person’s afternoon, and sold at the other, in units of currency. Goldhaber’s intrinsic scarcity is real, and it is exactly what makes the inventory valuable.

That matters for anyone thinking about what comes after money as a measure of a life. The most cited candidate replacement is already owned.

What a ranking that maximises attention selects for

If attention is the inventory, the sorting function is whatever gets more of it. We have a decade of measurements of what that function picks.

The clearest is a study of 563,312 social media messages about three polarising moral and political issues. Each additional moral-emotional word in a message increased its diffusion by about 20 per cent. The second finding is the one usually left out: “moral contagion was bounded by group membership; moral-emotional language increased diffusion more strongly within liberal and conservative networks, and less between them” (Brady, Wills, Jost, Tucker & Van Bavel 2017, PNAS).

Read that as a specification and it is unpleasant. The score pays best for heat, and it pays best for heat aimed at people who already agree with you about who the enemy is. This is correlational work on how content diffuses, so we hold the size of the effect loosely. The direction has held up across a lot of subsequent work, and we have made the wider case for it in The Seat Is Already Filling - With Attention rather than repeating it here.

How much of a life this now takes

The numbers are not small and they are not guesses. Japan’s Ministry of Internal Affairs and Communications has run an annual diary study of media time since 2012. In the survey published in June 2026, covering a diary week in December 2025 with 1,800 people aged 13 to 79, average weekday internet use was 183.9 minutes against 156.1 minutes of real-time television. On rest days the figures were 192.9 and 189.2 minutes. The ministry’s own summary notes that internet use has now passed real-time television viewing on both weekdays and rest days, and that the gap is widening (MIC 2026).

Three hours a day, every day, inside systems whose ranking function is the one described above. That is not a fringe of life. For most waking hours that are not work or sleep, it is the environment.

What it does to the people being counted

The best causal evidence we have on the last question comes from an experiment rather than a survey. Hunt Allcott, Luca Braghieri, Sarah Eichmeyer and Matthew Gentzkow paid a randomised group to deactivate Facebook for four weeks before the 2018 US midterm election. Deactivation “reduced online activity, while increasing offline activities such as watching TV alone and socializing with family and friends”, “reduced both factual news knowledge and political polarization”, “increased subjective well-being”, and “caused a large persistent reduction in post-experiment Facebook use.” Their conclusion about measurement is the sharpest part: deactivation lowered what people would later pay for the product, “suggesting that traditional metrics may overstate consumer surplus” (Allcott, Braghieri, Eichmeyer & Gentzkow 2020, American Economic Review).

We will not overclaim from one study. It is one platform, one country, one month, one election. It also found a real cost to leaving: people knew less about the news. But the finding that people value the thing less once they have been away from it for a month is hard to square with the idea that engagement measures what anyone actually wants.

Why it takes the seat when nothing else is installed

Here is the part we think is under-argued, including by us. The usual explanation for attention winning is that it is addictive, or that platforms designed it that way. Both are true and neither is the structural reason.

The structural reason is that attention is the only measure of standing that costs nothing to compute. A view is counted as a byproduct of delivering the thing. Nobody has to convene, deliberate, know your work, or take responsibility for a judgement. Every other way of honouring a person is expensive: a peer has to know what you did, decide it was good, and put their own name behind saying so.

Cheap measures beat expensive measures by default, the same way a free tool beats a good one in an organisation that has not decided which it wants. That is why we argue in There Is Always a Yardstick that the seat is never empty. If a society does not deliberately fund and build an expensive way of conferring standing, the cheap automatic one is what it gets, and it will not have chosen it.

None of this means attention is bad, or that being read is shameful. Attention is how anything gets known at all, this essay included. The claim is narrower. Attention is a terrible measure of a life because it was never designed to be a measure of anything. It is a counter that came free with the medium, and free things end up load-bearing.

Make impact, not money, the measure of a life. And not clicks either.

Sources

← All essays

Impactism is free — no paywall, no membership fee, and it will stay that way. If this argument was worth your time, donations are what pay for it. One-time or monthly, any amount.

Support the argument