Nobody can price something they have never measured. That sounds obvious in every other market, and it is the step almost everyone skips with attention. People will say their time online is worth something, then be unable to answer a simpler question: how much of it went out yesterday, and to whom?
Building a marketplace where people set a price for their attention kept running into that gap. The first version of the onboarding assumed the interesting decision was the number — pick a rate, publish it, start receiving paid messages. In practice the number was the last decision, not the first. The work that changed behaviour happened before it, in a plain audit most products are too impatient to ask for.
Designing the audit first
Three choices shaped this part of the build, and none of them involve a price.
First, the audit is descriptive, not judgmental. Which platforms got the longest sessions yesterday? Which content is still remembered an hour later, and which left nothing behind? Those two questions sort a day of scrolling into attention that returned something — learning, connection, rest, a direct reply — and attention that was simply collected. No score, no grade. The moment an audit feels like a report card, people stop filling it in honestly.
Second, limits come before pricing. Batched notifications instead of constant ones, one intentional session instead of many reflexive checks, a set window for feeds. In testing the sequence mattered: people who had already reclaimed some time priced the rest of it calmly. People asked to price attention they were still leaking everywhere either picked a fantasy number or gave up on the form.
Third, the price is a floor, not a valuation of the person. The working framing became blunt: below this amount, this exchange is not worth interrupting a day for. A floor can be raised later without embarrassment. A supposed total value of someone's attention cannot, which is why asking for one stalled people.
What worked
The audit changed habits before any money moved. Once attention started to feel like a budget rather than an endless supply, sessions got shorter on their own — not because a tool blocked anything, but because the spending had become visible. That is a cheaper mechanism than any blocker, and it survives uninstalling the tool.
Floors also filtered better than expected. Setting a price in advance changed the quality of what arrived: messages written to a person whose time has a stated cost are shorter, more specific, and far more likely to say what they actually want in the first line. Earners on the platform keep 85% of what they charge, and that split only matters if the messages reaching them are worth reading in the first place.
And the 18+ boundary, which looked like pure compliance, turned out to be a design asset. An adults-only marketplace can talk plainly about money, interruption, and refusal in a way a general-audience feed cannot, and that plainness is most of the product.
What broke, or nearly did
The first pricing screen broke immediately. It asked for a single rate, up front, with no context — and it read as a judgement on the person, not on an inbox. Completion recovered when the rate moved to the end of the flow, after the audit and the limits, where it answers a question the user has already been helped to think about.
Comparison nearly broke it next. Showing what other people charge feels like useful market information and behaves like poison: new users anchored to strangers' numbers, underpriced themselves to look approachable, then resented the messages that arrived at that price. Rates are now set privately, from the user's own audit, and comparisons stay out of onboarding entirely.
The honest limit: a floor does not create demand. Pricing attention makes an exchange fairer when someone wants that attention; it does not make the phone ring, and the build does users no favours by implying a rate card is an income. What it changes is the direction of the default — attention stops flowing out free and uncounted, whether or not any particular message ever gets paid for.
Lessons for builders
Measure before you monetise. If users cannot describe where a resource goes, a price field is just a guess with a currency symbol.
Sequence is a feature. The same three steps — audit, limits, price — failed in one order and worked in another. Onboarding order deserves the same testing budget as onboarding copy.
Keep self-worth out of the form. Price the interruption, the message, the hour — never the person. Users can negotiate a floor; they cannot negotiate a verdict.
Hide the leaderboard. Any number users set for themselves gets worse the moment they can see everyone else's.
The project behind these lessons is Introworth, a marketplace where people set the price of their own attention, choose which brands and creators they interact with, and keep 85% of what they charge — free to join and set a price, 18+.
