The new payment standard x402 enables AI-powered agents to make payments instantly.
Original text: Pricing the Internet
Author: Sumanth Neppalli and Nishil Ja
Compiled by: Plain Language Blockchain
Cover: Photo by Mehdi Mirzaie on Unsplash

In the crypto space, two distinct schools of thought exist. As a book, we have the opportunity to examine these two perspectives closely. One view holds that everything is a market , and that pricing things is the path to success. The other believes that crypto technology is a better financial technology infrastructure . Our publication calendar lies between these two viewpoints because, like all markets, there is no single truth. We are simply tracing all the possible models.
This article delves into a new standard of payment that is emerging online. Simply put, it poses the question: what would happen if you could pay per article? To find the answer, we look back to the early 1990s to see what happened when AOL tried pricing internet access. We examine how Microsoft priced its SaaS subscription service. Finally, we stop at a coffee shop and see Claude charging by the minute for conversations based on the amount of text.
In this process, we explained what x402 is , who the key players are , and what it means for platforms like Substack.
The business models of the internet are out of touch with how we use it.
In 2009, the average American visited over 100 websites per month. Now, the average user opens fewer than 30 apps per month, but spends significantly more time within those apps. Back then it was about half an hour a day; now it's close to five hours.
The winners—Amazon, Spotify, Netflix, Google, and Meta—became aggregators , bringing together consumers with demand and turning leisure into habits . They priced these habits as subscriptions.
This works because attention follows a pattern. We watch Netflix most nights. We order groceries from Amazon weekly. Prime plans include $139 a year for shipping, returns, and streaming. Subscriptions eliminate most of the ongoing pain . Amazon is now pushing ads at subscribers to extract profits, leaving users to endure ad spending or higher fees . When aggregators can no longer justify subscriptions, they'll revert to ads, just like Google Play, which monetized by discontinuing cancellations.
Let's see what's flowing through the pipe right now:
- Robots and automation now account for nearly half of all web traffic. This is largely thanks to the rapid adoption of artificial intelligence and large language models (LLM), which have made robot creation easier and more scalable.
- API requests account for 60% of the dynamic HTTP traffic processed by Cloudflare. In other words, machine-to-machine communication consumes the majority of the traffic.
We designed today's pricing model for the human-participatory internet. Now, traffic is machine-machine , and it's sporadic . Subscriptions change habits. Listen to Spotify on the way to work, use Slack during work hours, and watch Netflix at night. Ads pre-determine a personality—someone is scrolling, clicking, thinking. But there's no familiarity with machine-machine, no eye contact. They have robbery and tasks .
Content pricing is not only a function of market contracts but also of general distribution infrastructure. Music existed in album form because physical media required bundling. The cost of burning one song or twelve songs onto the same CD was roughly the same. Participants needed high profit margins, and booth space was limited. In 2003, when decentralized energy shifted to the internet, iTunes changed its unit of account to songs . You could buy any song from iTunes on your computer and then sync it to your iPod.
Unbinding increases discovery, but also outlines revenue . Most fans buy hit songs, not ten full tracks, which squeezes per-listener revenue for many artists.
Then, with the advent of the iPhone, the infrastructure changed again. Inexpensive cloud storage, 4G, and a global CDN made access to any song instant and seamless. With your phone always online, you had virtually unlimited access to a vast library of songs. Streaming at the access tier bundled everything back together: all the music for just $9.99 a month.
Currently, music subscriptions account for over 85% of music revenue. This is something Taylor Swift is unhappy about: she was forced to switch to Spotify.

Enterprise software follows the same logic. The product is digitized, and the vendor can charge for the resources they intend to use . B2B SaaS vendors typically offer predictable monthly or yearly service access "per seat" and set policies that limit functionality, such as $50/user/month plus $0.001/API call.
The subscription topic describes predictable human use and measures bursts of processing machine usage .

When AWS Lambda runs your functions, you need to be accurate about the cost of the resources you consume . B2B transactions typically involve bulk orders or high-value purchases, resulting in larger transaction volumes and significant recurring revenue from smaller, more concentrated user groups. Last year, B2B SaaS revenue reached $500 billion, twenty times the size of the music streaming industry.
If most consumption is now machine-driven and sporadic, why are we still pricing it the way we did in 2013? Because our current infrastructure is designed for humans to make choices occasionally . Subscriptions become the default option because a monthly decision is better than a thousand small payments .
This is not simple, because encryption technology has created the infrastructure that now supports small payments. While that's one reason, the internet itself has become a behemoth , requiring new ways to use it for pricing.
Why did the micropayment fail?
The dream of paying a few cents for content is as old as the internet itself. Digital equipment company Millicent's protocol promised transactions for less than a cent in the 1990s. Chaum's DigiCash took banking action. Rivest's article PayWord solved the problem of what people were learning. Every few years, the same elegant idea is rediscovered: what if you could pay $0.002 per article, $0.01 per song—exactly the cost?
They all suffered the same failure: humans hate measuring their enjoyment.
AOL learned this at a great cost back in 1995.

Source - Cases Against Micropayments
They charged for dial-up internet access by the hour. For most users, this survey showed it was cheaper than a monthly subscription. However, customers hated it because of the mental burden . Every minute online felt like a running meter, every click incurring a tiny cost. People unconsciously accumulated each small cost into a "loss," even if the amount was small. Every click became a tiny decision: Is this link worth $0.03?
In 1996, when AOL switched to unlimited service , usage tripled overnight.
People pay more to think less. "Paying precisely for what you use" sounds efficient, but for humans, it often feels like a price tag tinged with anxiety .
In his 2003 paper, "The Case Against Micropayments," Odlyzko concluded that people pay more for fixed-rate plans not because they are rational, but because they crave predictability over efficiency . We'd rather pay an extra $30/month for Netflix than optimize every $0.99 title. Later attempts, such as Blendle and Google One Pass , tried to increase the price from $0.25 to $0.99 per article, but ultimately failed. Unit economics don't work unless a large percentage of the readership is converted, and the user experience increases cognitive costs .
Subscribe to Hell

What is life? Isn't it just continuous suffering? Perhaps the gods also have a subscription model for the existence of humanity.
If we crave the simplicity of subscriptions, why are we complaining about subscription hell today? A simple way to think about pricing is to ask yourself how often you experience the pain of using the product.
Entertainment demands are limitless. This black line represents a persistent pain point, a dream for both users and companies—a predictable, continuous curve of suffering. This is why Netflix transformed from a quirky DVD-writing service into a member of the elite FAANG club, offering an unlimited library of content and eliminating billing fatigue .
The simplicity of subscriptions has permeated the entire entertainment industry. As Hollywood studios saw Netflix's stock price soar, they began to reclaim their own content catalogs and build their own subscription empires: Disney+, HBO Max, Paramount+, Peacock, Apple TV+, Lionsgate, and more.
The fragmentation of content libraries leads users to purchase more subscriptions . If you want to watch anime, you need to subscribe to Crunchyroll. For Pixar movies, you need to subscribe to Disney. Watching content becomes a portfolio-building issue for users.
Pricing depends on two things: how accurately the underlying infrastructure can measure and bill usage , and who must decide when value is consumed each time .
One-time donations are very effective for rare, high-profile events. Buy a movie. Pay a one-time consultation fee. The pain strikes intensely once, then disappears. This pattern works when the task is neither meaningful nor valuable. In some cases, this pain is even desirable. The shopping experience romanticizes our trip to the theater or bookstore.
Precisely measuring usage ensures pricing stays close to the unit of work . That's why you won't be a half-moon payer. The value there is vague. Figma can't extract a fixed portion from your monthly stairwell; the value created is difficult to quantify .
Even if it's not the most profitable method, prepaid monthly fees are easy to shave off.
Computation is different: the cloud can observe every millisecond. Once AWS can measure execution at this granularity, renting an entire server becomes pointless. Servers are only started when needed, and you only pay for what you're doing. Twilio did the same for the telecom industry: one API call, one SMS fragment, one charge.
Ironically, where we could perfectly measure, we still measure the same way. We run meters on a monthly basis, but the cash flow is through monthly credit card subscriptions, PDF invoices, or prepaid “credit lines.” To achieve this, you subject each vendor to the same ordeal—creating an account, setting up OAuth/SSO, granting API permissions for authorization, storing a card, setting a monthly cap, and ensuring you don’t overcharge.
You preload credit limits using tools. Others, like Claude, will limit you to a lower model when losses are reached.
Most SaaS providers have a green "predictable load" range. This is too congested for short-term purchases, yet too stable to justify accurate event-based metering. The strategy here is bucketing and tiering . You choose a plan that suits your typical monthly usage and upgrade when you exceed the limit.
Microsoft's "1TB per user" is one such example, differentiating users from heavy users without counting every file operation. The CFO limits the number of users who need access to higher-level systems by assigning permissions.

Chaotic Middle Ground
A simple way to organize a pricing model is with a dual-axis graph , where the x-axis represents usage frequency and the y-axis represents usage variability . Here, variability means sporadic occurrence —how irregular an individual user's usage patterns are over time. Watching two hours of Netflix most nights is low variability; an AI agent making 800 API calls in ten seconds, then going quiet and then calming down, is high variability.

In the bottom left corner is the one-time purchase option. Simple "buy and go" pricing works when tasks are rare and predictable, because you only need to pay once and then move on.
The top left corner is a cluttered, casual webpage with irregular news browsing, link jumps, and a low willingness to pay. Subscriptions are redundant, and pay-per-click micropayment decisions and transaction frictions crumble. Advertising becomes the financing layer, aggregating millions of opinions with varying variance. Global advertising revenue has surpassed the $1 trillion mark, with digital advertising allocating 70% of the spending, indicating that most of the web is in this low-commitment corner.
The bottom right corner is where subscription habits make a lot of sense. Slack, Netflix, and Spotify align with human daily routines. Most SaaS products are located here, with tiers separating users from isolated users. Most products offer a freemium tier to encourage users to start using their product, then gradually shift their usage from the top left to the bottom right corner through daily, consistent habits.
Subscriptions account for approximately $500 billion in global annual revenue.
The top right corner represents the core of the modern internet , encompassing LLM queries, intelligent agent actions, serverless bursts, API calls, cross-chain transactions, batch jobs, and IoT device communication. Usage is continuous and irregular . The fixed fee based on seats misprices this reality but lowers the psychological barrier to starting to pay. Due to an overabundance of paying users and a shortage of heavy users, revenue interrupts actual spending .
This is why seat-based products tend to favor metering . A basic plan is reserved for collaboration and support, but heavy loads are charged. For example, Dune offers a limited monthly credit limit . Small, simple queries are inexpensive, while longer-running queries require more credit.


If spending is trending towards consumption close to the pound , the market tells us that pricing is hoping to keep pace with work. Machines are rapidly becoming the largest consumers of the internet, with half of all consumers using AI-driven searches. And now, machines are creating more content than humans.
The problem is that our infrastructure still relies on annual accounts . Once you successfully register for the software, you can access their dashboard, which includes API keys, prepaid credit lines, and end-of-month invoices. These work fine for habitual humans; but they become cumbersome for bursty software. In theory, you could use ACH, UPI, or Venmo for regular monthly reports. However, these require batch processing , and their cost structures crumble under low-cent, high-frequency traffic .
This is the significance of encryption in the internet economy . Currencies provide you with instant, stable, and granular payment capabilities down to the cent . They operate internally for settlement and alerting, rather than being held directly by intelligent agents or trapped behind a bank's user interface. If usage should be event-driven , so should settlement, and encryption is the first infrastructure capable of keeping up with this reality .
What exactly is x402?
x402 is a payment method that works with HTTP and uses the 402 status code , which has been a standard feature for small-value payment devices for many years.
x402 is a way for sellers to verify that a transaction has been completed . Sellers who wish to accept on-chain, gas-free payments via x402 must work with coordinators (service providers) such as Coinbase and Thirdweb.
Imagine Substack raises the price of a paid article by $0.50. When you click the "Pay to Read" button, Substack returns a 402 code containing the price, accepted assets (such as USDC), networks (such as Base or Solana), and strategy. It looks something like this:

Your Metamask wallet authorizes $0.50 via a signed message to submit it to the coordinator. The coordinator sends the text on-chain and notifies Substack to open the article.
Stablecoins make accounting fast . They settle at network speed , in tiny denominations , and are consumed in accounts at each provider. With x402, you consume pre-funded in five credit buckets , consumption rotates API keys in different environments, and there's no risk of being caught in a mid-process trigger at 4:00 AM, causing your jobs to be interrupted. Bills can remain where they're best suited to them—on the billing statement—while all spikes, machine-to-machine paths are handled automatically and cheaply.

You can achieve this difference in agent checkout . Let's say you're experimenting with a new fashion style on Daydream (an AI fashion chatbot). Today, the shopping process redirects you to Amazon so you can pay using your saved card details. In the world of x402, the agent understands the context, retrieves the merchant's address, and pays from your Metamask wallet without leaving the current conversation thread.

The interesting part about x402 is that it's not a single entity at the moment; it's composed of various systems you'd expect to find in real-world infrastructure. Anyone building AI agents through Cloudflare's Agent Kit can create action-based pricing bots. Regional payment providers like Visa and PayPal are also adding x402 as part of their supporting infrastructure.
QuickNode has a best practice guide on how to add an x402 paywall to any endpoint. The direction is clear: unify "agent checkout" at the SDK layer, making x402 the way to make payments, tools, and ultimately retail purchases for agent APIs .

Integrating x402
Once the network supports rebuilding payments, the biggest questions will arise first. The answer is high-frequency usage areas , where transactions are worth less than $1 . That's where subscriptions are overpriced for subscribers. These monthly commitments require subscribers to pay a minimum subscription amount to even begin using the service. As long as blockchain fees are feasible, x402 can settle each request at machine settlement speeds, roughly as low as $0.01 .

Two forces are making this sense of urgency urgent. On the supply side, the tokenization of work is exploding—LLM tokens , API calls , search services , IoT ping . Every meaningful action on the modern network now comes with a tiny, machine-inevitable unit . On the demand side, SaaS pricing leads to absurd waste. About one-tenth of licenses are idle because teams prefer to pay per seat, as it's easier to monitor and predict. We measure work at the technical level, but pay humans at the seat level.

Limiting event intervals is one way to keep these markets aligned without scaring buyers. We can set a cap on software to run at the optimal price. A news website or developer API might charge per request all day, then automatically refund to the daily cap.
If The Economist publishes an article for $0.02 per article, with a maximum of $2 per day , an interested reader can browse 180 links, and the cost is calculated mentally.
At midnight, the agreement settles all fees to $2. The same model applies to the developer interface. News organizations can charge per LLM crawl to sustain future AI browser revenue. Search APIs like Algolia can charge $0.0008 per query, aggregating daily usage to $3.

You can already see consumer AI moving in this direction . When you reach Claude's messaging limit, it won't just say, "Limit reached, try again next week." It will offer two paths on the same screen: upgrade to a higher subscription , or pay per message to finish what you're doing.
A simple infrastructure is needed to allow the agent to automatically make a second choice, pay on request , and then pop up an interface for the user to use a button or upgrade manually.
For most B2B tools, the actual end state looks like a " subscription base price + x402 burst ". The team maintains a base plan tied to the number of people for collaboration, support, and backend use. Occasionally, there are recalculations (build minutes, search provisioning, image generation) billing via x402 traffic, which forces an upgrade to the next system.
Better networks are also available. DoubleZero aims to sell faster, clearer internet through dedicated devices. By routing agent traffic through these devices, you can price your x402 service in gigabytes with a clear SLA and cap. An agent requiring low latency for transactions, rendering, or model switching can enter a period of temporarily fast bandwidth, pay for that specific burst, and then exit.
SaaS will accelerate the shift to usage-based pricing, but with accompanying safeguards :
- It's cheaper for you to acquire and activate integration. You can earn money from the first call. Even if a developer never completes an OAuth or credit card form, they can still be paid $0.03. Agents tend to favor providers that offer instant payments .
- Revenue grows as the number of seats actually used expands. This is how to address the 30-50% seat waste in most organizations. Heavy workloads are shifted to top-level bursts .
- Pricing becomes a product interface. "Fast lane, an extra $0.002 per request" and "Bulk mode half price increase" are examples of revenue-generating outlets companies can try.
- Lock in the split. This reduces conversion costs by consuming integration work and time, leveraging the supplier's capabilities.
A world without ads
Small payments won't eliminate advertising entirely; they will narrow the areas where advertising is the only viable model . Advertising will still perform well in leisure real estate spaces. x402 prices for interfaces where advertising is inaccessible and for situations where someone might occasionally choose to pay for a great article , thus consuming a monthly subscription.
The x402 simplifies payments ; on a certain scale, it could change the future .

With 50 million users and a 10% conversion rate, Substack has 5 million subscribers, each paying approximately $7 per month. When the paid subscriber base doubles to 10 million, Substack could potentially see even more additional payments in its tiered subscription model . Because the conversion rate is relatively low, more casual readers can switch to pay-per-view , accelerating the revenue curve.
The logic applies to any seller with highly differentiated, low-frequency sales : when people use a product occasionally and differently , pay-as-you-go habits feel more natural than a commitment to a long-term plan.
It's a bit like going to my local badminton court. I play two or three times a week, usually in different places with different friends. Most courts offer monthly memberships, but I prefer not to tie myself to one place. I like the freedom to decide which court we go to, how many we go to, and whether I can skip a lesson when I'm tired.
Don't get me wrong, I know it varies from person to person. Some people like to consistently go to the nearest course, some like to subscribe to motivate themselves to form a habit, and some might want to share a subscription with a friend.
I can't speak for offline payments, but with the X402, this personalization can be digitally realized. Users can set their own payment preferences through strategies, and companies can provide flexible pricing models to suit everyone's habits and choices.
The x402 truly shines in its agent workflow . If the past decade was about turning humans into logged-in users , the next decade will be about turning agents into paying customers .
We're halfway there. AI routers like Huggingface allow you to choose from multiple LLMs. OpenAI's Atlas is an AI browser that uses LLMs to run tasks for you. x402 plugs into that world as an offloaded payment infrastructure. It's a way to let software settle small bills with other software at the final moment when the work is done .
However, simply having infrastructure doesn't always create a market. Web2 built its entire scaffold around card networks. There's KYC for banks, PCI for merchants, PayPal disputes, suspensions due to fraud, and refunds when problems arise. Agent commerce currently lacks these. Stablecoins plus HTTP 402 statements offer agents a payment method, but they also disrupt the built-in recourse that people are used to .
How do you get your money back when your smart shopping purchases the wrong flight or your research bot goes over budget?
This is the question we will explore in more depth next: how developers can use x402 without worrying about future failures.
Disclaimer: As a blockchain information platform, the articles published on this site represent only the personal views of the authors and guests and do not reflect the position of Web3Caff. The information contained in the articles is for reference only and does not constitute any investment advice or offer. Please comply with the relevant laws and regulations of your country or region.
Welcome to the official Web3Caff community : Twitter account | Web3Caff Research Twitter account | WeChat reader group | WeChat official account




