I hit a paywall this morning.
While I was researching for a recent project, I found an article with a perfect headline and a compelling thesis teaser. But to read the full analysis, I was asked to sign up and pay for a $200 annual subscription. Valuing my time and my money, I did what most people do: I closed the tab.
On the surface, this was a conversion failure, caused by the payment friction from the checkout process and payment cost structure. However, there is a deeper issue: a fundamental mismatch between how we consume digital value and how we pay for it. We are entering the era of the Evaluation Economy, and our legacy payment rails aren't ready.
The Information Asymmetry Trap
The reason I didn't click "subscribe" wasn't necessarily the price point; it was the risk. I was being asked to price a product before I could assess its value. Economists call this information asymmetry: it occurs when the seller possesses more knowledge about a product's quality than the buyer. It is not an issue when we buy a smartphone or a laptop, since we can thoroughly evaluate them via specs and reviews (aka search goods), and it is perfectly rational to pay upfront. However, the quality of research paper or a proprietary data source cannot be known before consumption (aka experience goods), and it makes paying upfront risky. Even worse, information is not returnable: I cannot “un-read” a paper or “return” a piece of data, which makes refund unrealistic.
To understand why our current models are failing, we have to look at how products sit on the intersection of Information Asymmetry and Consumption Patterns.
- Low information asymmetry, discrete consumption: standardized transactions such as consumer electronics and airline tickets. Product quality is largely known ex-ante, and consumption occurs in discrete units.
- Low information asymmetry, continuous consumption: recurring utility services such as streaming platforms and cloud infrastructure. Value is predictable, and consumption occurs over time.
- High information asymmetry, discrete consumption (upfront risk purchases): products such as online courses, books, and research reports. Value is difficult to assess prior to purchase.
- High information asymmetry, continuous consumption (evaluative consumption): categories such as digital media, AI services, and data products. Value is revealed progressively through usage.
The "Subscription Paradox"
In low information asymmetry markets, a limited trial access usually provides a reliable signal of overall product quality to consumers. The content available during a free trial of a streaming service is very likely consistent with the full paid offering, so users can have a high confidence in their expectations of full product quality. This makes freemium and trial-based models sufficient for monetization.
This breaks down in highly asymmetric markets. For products such as online courses or video games, samples are often curated and unrepresentative. A preview chapter or game demo provides only a weak signal of overall quality. In these cases, freemium is insufficient.
Products that require continuous consumption such as digital media and data streams face additional transaction challenges.
First, each transaction carries a fixed cost. Traditional card rails have fixed per-transaction fees, making small, high-frequency payments uneconomical.
Second, payments require user authorization. Each transaction is a discrete interruption, introducing friction that prevents payment from aligning with continuous consumption.
To solve these challenges, platforms default to subscriptions that bundle access, effectively shifting risk to the consumer. Subscriptions consolidate many small interactions into a single payment, avoiding both transaction costs and repeated authorization.
But this comes at a cost: subscriptions reduce payment friction, but they further increase the information asymmetry. Subscription qualities are difficult to assess effectively. While high-quality writers face higher costs (e.g. paying for professional data) and must charge accordingly, lower-quality newsletters, with lower costs (e.g. written by AI), can undercut on price. The subscription paywall weakens price signals, makes it harder for high-quality writers to differentiate, and lowers readers’ willingness to pay, leading to a classic “lemons problem”.
Subscriptions, in this sense, are an imperfect compromise between continuous consumption and discrete payment systems. Solving this requires a shift not just in how we price products, but in the fundamental architecture of how payments are executed.
Machine Payments: From Transactions to Flows
To move beyond subscription, we must recognize that our current payment architecture is fundamentally discrete. It was designed for events that have a clear start, a fixed price, and explicit human authorization. Digital value is increasingly continuous. We need a transition to machine payments, a paradigm where transactions are low-cost, programmable, and automated.
1. Solving Continuous Consumption: “Pay as you Scroll”
Legacy rails force us to bundle content because individual transactions are too expensive and disruptive. Machine payments allow for on-the-fly monetization.
Imagine a reader scrolling through a research report. Instead of an upfront $200 commitment, the writer charges for the article in real-time. As the reader moves through the text, a stream of micro-payments is executed in the background. They can even pay to unlock a specific data table or a supplementary video deep-dive that appears midway through an intriguing section. The consumption and the transaction are finally in sync.
2. Solving Information Asymmetry: Real-Time Evaluation
Because these payments happen in real time, the information asymmetry is eliminated. This is evaluative consumption: the consumer assesses the product and determines the price during consumption.
You are no longer guessing if a publication is worth an annual fee. You only pay for the value as it is revealed. If the article proves irrelevant, you stop reading, and the payment stops.
This is made possible by three converging technologies:
- Stablecoins and real-time settlement: To settle a $0.05 micro-payment for a single paragraph, the cost of the transaction must be minimal. Stablecoins on modern rails provide the high-frequency infrastructure that legacy card networks, with their heavy overhead, cannot match.
- Programmable logic: Payments are no longer static. They are governed by code that pays $0.01 for every 10% of the video watched or $0.50 when the data table is expanded.
- Agentic authorization: We cannot ask a human to click "confirm" every time they scroll a page. AI agents manage pre-set budgets and preferences, executing micro-transactions in the background so the experience is frictionless.
By shifting from discrete transactions to continuous flows, we eliminate the “lemons problem”. We move from a world where we pay for access to a world where we pay for proven value.
The $5 Trillion Payoff
The implications of this shift extend beyond simple efficiency. According to McKinsey, agentic commerce could drive up to $5 trillion in global revenue by 2030.
Under traditional models, demand is artificially constrained by that upfront commitment. When you force users to commit before value is revealed, the willingness to pay is structurally lowered. By enabling continuous payment, conversion rates will be higher, stickiness will increase (because of more frequent engagement), and high-quality long-tail products will have a chance to succeed by charging a fair price.
The transition to the evaluation economy marks a shift from gated access to fluid value. For decades, the industry has focused on making checkout faster; in the era of machine payments, the goal is for the checkout to disappear entirely.
Legacy payment rails, with their high fees and slow settlement, were never designed for the "pay-as-you-scroll" world. This is where stablecoins shine. By providing a programmable, real-time settlement layer, stablecoins allow value to move at the same speed as data, enabling micro-transactions that are finally economically viable.
As Jack Forestell, Chief Product and Strategy Officer at Visa, noted, this is the single biggest opportunity in 20 years of payment technology. It is the moment blockchain becomes the invisible engine of global commerce. We are no longer asking users to pay for the promise of value; we are building the rails for them to pay for the reality of it, one micro-settlement at a time.
.png)
