What's the Simplest Way to Accept Payments Online? Why Fewer Features Mean Fewer Problems

by PayFi · 5 October 2026 · Notes from building

Payment tools compete by adding features: accounts, dashboards, mailing lists, wallet pop-ups, balances, loyalty points. Each one sounds helpful. Each one also adds setup work, another step for buyers, more data to protect and more things that can break. Sometimes the most useful thing a payment tool can do for you is leave things out on purpose.

A weekend lost to setup

Let's meet a seller we will call Rina. She is not a real person, but her weekend is a familiar one.

Rina designs knitting patterns and sells them as PDF files. A few people on social media have asked to buy them. So one Saturday, she sets out to "do it properly".

She signs up for an online store builder. She picks a theme, then discovers she needs an app to deliver digital files, another for a mailing list pop-up, and another for discount codes. She connects a payment provider, which asks for her ID, her bank details and her tax number. She writes a privacy policy, because the store now collects buyers' names and emails. She sets up shipping settings for products she will never ship, because the store will not let her skip them.

By Sunday evening, she has a beautiful store, four monthly subscriptions, and no sales. Her first buyer messages her the next day: "Is there an easier way to pay? Your checkout wants me to create an account."

Rina's mistake was not laziness or bad taste. It was a belief most of us share: that a better tool is one with more features. This essay is about why that belief often works against small sellers, and about a different way to think about it.

Most features go unused

Software companies know something users rarely hear: most of what they build is hardly touched.

Pendo, a company that measures how people use software products, reported in 2019 that "80 percent of features in the average software product are rarely or never used" (Pendo). It estimated that public cloud software companies had collectively spent up to 29.5 billion dollars developing features that went unadopted or underused (Pendo).

Unused features are not free just because nobody uses them. Someone built them, someone maintains them, and the price of the product pays for them. In payment tools, they also tend to come with settings to configure, data to store and steps to click through.

Every extra step costs buyers too

Features are not only a burden for sellers. Many of them land on the buyer, as extra steps between "I want this" and "I've paid".

The Baymard Institute, which studies online checkouts, has measured this for years. In its survey of US online shoppers, 18 percent said they had abandoned a purchase because the site wanted them to create an account, and 17 percent because the checkout was too long or complicated (Baymard Institute). Baymard also found that the average checkout shows 23.48 form elements by default, and estimates that a good checkout could get by with about half of that (Baymard Institute).

Every one of those fields was a feature to someone: a newsletter opt-in, a phone number for delivery updates, a company name for invoices. For a buyer who just wants a PDF, each one is a small reason to leave.

What a feature really costs

It helps to look at a few common features and list what each one quietly brings along.

An account system brings passwords to store safely, reset emails to send, people locked out who need help, and a database of logins that attackers would love to have.

A mailing list at checkout brings an email field for every buyer, consent rules in many countries, a privacy policy to keep up to date, and a list of personal details to protect.

A customer balance brings payout schedules, reconciliation, rules for holding money, and the risk that the money is frozen or lost.

Identity checks bring document uploads, manual reviews, storage of passports and selfies, and the people who give up halfway.

Card payments bring fraud checks, disputes, dispute fees, and the possibility that a sale is reversed months later.

None of these costs shows up on a feature list. They show up later, in fees, in setup time, in support emails and, at worst, in a data breach. IBM's 2025 report put the average cost of a data breach at 4.44 million dollars worldwide and 10.22 million in the United States (CyberScoop). A small seller will never face a bill like that alone, but the services they use are paying for that risk, and pass the cost on.

The old idea: more is better

The belief behind Rina's weekend goes like this. A serious seller needs a serious setup. Serious setups have accounts, dashboards, mailing lists, analytics and integrations. If a tool lacks those things, it must be a toy.

That belief makes sense for some businesses. A shop with thousands of products, subscriptions, warehouses and staff really does need much of that machinery.

But for many sellers, especially people selling digital products, services or one-off work to buyers around the world, the machinery is mostly cost. What they need is narrower: a way to show a price, receive the money, deliver the product and prove the payment happened.

The rest of this essay walks through things PayFi decided not to build, and why. Not because they are bad ideas in general, but because each one would have added more cost, risk or friction for sellers and buyers than it removed.

Not built: accounts and passwords

Most payment services begin with "create an account". PayFi does not.

Sellers create a payment link by entering a wallet address, a product name, a price and a way to deliver the product. When they want to see their sales in the dashboard, they sign in by signing a short message with their own wallet, not with a username and password.

Buyers do not sign up at all. They open the link, see the product and the price, and pay.

What you give up: password reset by email, and a single profile that remembers you across purchases.

What you gain: nothing to remember and nothing to leak. There is no database of usernames, emails and passwords for anyone to steal, and no account for anyone to lock you out of. For buyers, it removes the step that Baymard's survey found makes almost one in five shoppers give up.

Not built: balances and payouts

Most payment services collect your sales into a balance and pay you later. PayFi does not keep a balance for anyone.

Each payment arrives at an address made only for that order. Within about 20 seconds, it is split in one transaction: 99 percent to the seller's wallet, 1 percent to PayFi.

What you give up: the feeling of a single account where all your earnings sit together, managed by someone else.

What you gain: there is nothing to freeze, nothing to hold for review, and no payout schedule to wait for. Your money is in your wallet, where only you control it. You can still see your sales in one place: the seller dashboard lists them, and your wallet's own history shows every payment that arrived. The difference is that these are records of money you already have, not a promise of money someone will send you later.

Not built: identity checks

PayFi does not ask sellers or buyers for an ID, a selfie or proof of address.

What you give up: the kind of protection that comes from a regulated company knowing who everyone is. Identity checks still happen where regular money enters and leaves the system, at the exchanges where people buy and sell digital dollars.

What you gain: you can start selling in minutes, and people without the right documents are not shut out. There are no copies of your passport sitting in yet another database.

Not built: card payments and chargebacks

PayFi does not process card payments. It only moves USDT and USDC on the Solana network.

What you give up: buyers who only have a bank card cannot pay directly. They first need to buy digital dollars, through an exchange or a card service, which costs them a fee the first time. That is a real cost, and it is the biggest reason some buyers will not use a payment link like this.

What you gain: no chargebacks. A card payment can be disputed and reversed months later, and services charge sellers for each dispute; Stripe, for example, charges 15 dollars for each dispute received (Stripe). A completed stablecoin payment is final. For sellers of digital products, who cannot take a file back once it is downloaded, that difference matters a great deal.

Not built: mailing lists and buyer data

Many checkouts ask for an email address, often with a box to tick for news and offers. PayFi does not ask for a buyer's name or email at all.

What you give up: a built-in way to collect buyers' contact details for marketing.

What you gain: buyers can pay without handing over personal information, and sellers do not become responsible for protecting a list of customers' details. According to its checkout page, the only things PayFi keeps for an order beyond the public payment data are a scrambled form of the buyer's IP address and the encrypted product link, and both are deleted automatically after 90 days.

If a seller wants a mailing list, they can still build one their own way, with buyers who choose to join. The payment does not force it.

Not built: "connect your wallet"

Many crypto websites start by asking you to connect your wallet and approve something. PayFi's checkout never does.

A buyer simply sends USDT or USDC to the order's payment address, from an exchange or from their own wallet, the same way they would send money to anyone. There is nothing to connect, and nothing to approve beyond the payment itself, confirmed in the buyer's own wallet or exchange.

This matters because connecting wallets and signing approvals is exactly how many crypto thefts happen. According to the security firm Scam Sniffer, as reported by BleepingComputer, so-called wallet drainers stole 494 million dollars from more than 300,000 wallet addresses in 2024, mostly by tricking people into signing approvals that let the attacker move their tokens (BleepingComputer).

A checkout that never asks buyers to connect a wallet or grant access to their tokens gives them one simple rule: if a payment page asks you to do that, you are not on PayFi.

Not built: pasting a transaction code

Some crypto payment pages ask buyers to send money and then paste the transaction code, the TXID, into a form so the seller can find the payment. PayFi does not.

Because each order has its own payment address, the system knows which order a payment belongs to as soon as it arrives. The page updates by itself. If a little less arrived than the price, for example because the buyer's exchange took a fee, the page says how much is still missing.

What you gain: one less fiddly step, and one less way for a buyer to make a mistake.

Not built: a coin of its own, or interest on deposits

PayFi has no token of its own to buy, hold or speculate on. It does not offer interest, rewards or staking. It moves two digital dollars, USDT and USDC, and that is all.

What you give up: nothing you need to get paid.

What you gain: no reason to wonder whether the service exists to sell you a coin. And because it holds no deposits, it has nothing to lend out or invest, which is how several crypto platforms that promised interest got into trouble in the past.

Not built: subscriptions

PayFi does not handle recurring payments. Each link is for a one-time payment.

PayFi's own home page says as much: where a platform already has its own subscription payments, PayFi does not replace them; it complements them for one-time products.

What you give up: automatic monthly billing.

What you gain: a buyer never has to worry about being charged again without noticing. Each payment is a deliberate act.

Not built: an app to install

PayFi has no app for buyers or sellers to download. Sellers create links on a web page. Buyers pay on a web page, from any phone or computer, using whatever exchange or wallet they already have.

What you give up: notifications and shortcuts that an installed app might offer. Sellers who want instant alerts can connect a Telegram bot instead, which sends a message when a sale is paid.

What you gain: nothing to install, update or give permissions to. A buyer who has never heard of PayFi can pay from the phone in their hand, in the browser they already use.

Not built: shipping and addresses

The checkout has no fields for a name, street address or postcode. PayFi's delivery options are all digital: a file link, a redirect to a private page, a note, or a receipt for a service.

What you give up: a ready-made way to collect delivery addresses for physical goods. A seller of physical products would need to arrange the address separately, for example in the chat where the buyer found them.

What you gain: buyers of digital products are never asked for a home address they do not need to give. For the great majority of things sold this way, files, access, services and invoices, an address field would only be one more reason to stop.

What is left

Take all of that away, and what remains is small, but it covers what most sellers of digital products and services actually need.

A link. Created in minutes, with a wallet address, a product name and a price. It can go in a bio, a chat, an invoice or an email.

A checkout that explains itself. The buyer sees the product, the price and the network to use, saves a private order link, and gets a payment address for that order only.

Delivery. When the full price arrives, the product unlocks on the buyer's page: a file link, a redirect to a private page, a note, or a receipt for services.

Your money in your wallet. Split within about 20 seconds, 99 percent to you.

Refunds by rule. Paid too much, too little, too late or the wrong coin? The buyer can get it back from their order page, minus a small fee, without asking anyone.

A public record. Every payment is a Solana transaction anyone can check, and a line in a public ledger that is sealed into Bitcoin.

That is the whole product. It is not trying to be a store builder, a bank, a marketing platform or an exchange.

The buyer's side, counted

It is worth counting what Rina's Canadian buyer actually did, because that is where a short feature list shows its value.

She opened the link and saw the pattern's name and price. She ticked a box confirming she would send on the Solana network, then saved her private order link, as the page asked. She pressed a button to show the payment address. She switched to her exchange app, pasted the address, typed the amount and sent it. A minute later, the page told her the payment had arrived, and the pattern opened.

That is the whole journey. She did not invent a password, confirm an email, type a shipping address, decline a newsletter, connect a wallet or paste a transaction code.

Compare that with the store Rina had built over the weekend, where the first step was "create an account". By Baymard's survey, that one step alone is enough to make almost one in five shoppers leave. Removing it was not a small improvement. For some buyers, it was the difference between paying and closing the tab.

What was built instead

Leaving things out does not mean building nothing. The effort that would have gone into accounts, balances and mailing lists went into a few things most users never see, but that protect them.

Automatic refunds. A small program checks a refund queue every 15 seconds and sends money back by rule, without anyone approving it.

The public ledger. Every payment is written into a book anyone can read, with a summary that should show nothing unaccounted for.

Seals in Bitcoin. The ledger's fingerprint is stamped into Bitcoin at most once an hour when there are new lines, and at least once a day, so the past cannot be quietly rewritten.

A checker anyone can run. A small program, offered on the ledger page, that verifies the book against Solana and Bitcoin directly, without asking PayFi.

These are not features you click on. They are features that make it unnecessary to trust the service blindly, which is a better use of effort than another dashboard.

Saying no is harder than saying yes

There is a reason most products keep growing. Every new feature has someone who wants it, and saying yes feels generous. Saying no feels stubborn, even when it is right.

A payment tool that wants to stay small has to judge each new idea against the beliefs it started with. Does this make the buyer's path longer? Does it mean holding money, or holding data about people? Does it make the system harder to check? If the answer is yes, the idea has to be very valuable to be worth it.

For sellers, it can help to apply the same discipline to their own setup. Every app, plugin and pop-up was a yes at some point. Some of them are still earning their place. Some are just costing money and buyers.

Who this is not for

Leaving things out means some people are better served elsewhere, and it is more useful to say so than to pretend otherwise.

Businesses that need subscriptions. Monthly memberships and recurring billing need a different tool.

Buyers who want card protection. If you value being able to dispute a charge through your bank, a card payment gives you that, and a stablecoin payment does not.

Sellers who need formal invoices with names and tax numbers. You can issue those yourself alongside a payment link, but the link will not produce them for you.

Shops with complex catalogues. Thousands of products, inventory, shipping rates and discount rules are what store builders are for.

People who want someone to call. A small team with automatic refunds and public records is designed so you rarely need support, but there is no large team on the phone.

If you are in one of these groups, the "missing" features are not missing for you. They are exactly what you need, and you should choose a tool that has them.

Why leave things out on purpose?

It is tempting to see a short feature list as a sign of a young product that has not caught up yet. Sometimes that is true. Here, it is a choice, and it follows from a few beliefs about what a payment tool owes the people who use it.

What you do not collect, you cannot leak. Every account, email list and identity file is something that has to be protected forever. Not collecting it is the strongest protection there is.

What you do not hold, you cannot freeze or lose. A service with no balances cannot lock your money in one, and cannot lose it in a collapse.

Every step you add, some buyers will not take. The shortest path from wanting to paying is the one that loses the fewest sales.

What is small can be checked. A tool that does a few things, in public, can be understood and verified by the people who use it. A sprawling platform mostly asks to be trusted.

What is not built does not need to be paid for. Fewer parts mean lower costs, and lower costs are how a flat 1 percent fee stays possible.

None of these beliefs is new. But together they explain why "not built" can be a feature in its own right.

Simple is easier to trust

There is one more benefit of a short feature list that is easy to overlook: people can understand it.

A buyer who meets an unfamiliar checkout wants to know three things. What am I paying for? Where is my money going? What happens if something goes wrong? On a page with accounts, pop-ups, wallet connections and fine print, those answers are buried. On a page that shows a product, a price, an address for this order and a written refund rule, they are in plain sight.

The same is true for sellers. When the whole system fits in a few sentences, it is easier to notice if something does not behave as described, and easier to check it on a public record. Complexity is not only a cost. It is also a place where problems hide.

How to choose for yourself

Whether you end up with PayFi or with a full store builder, a few questions help separate what you need from what you are being sold.

What do I actually sell, and how often? One-time digital products, services and invoices need far less than a catalogue with shipping.

What does my buyer need to do before paying? Count the steps and fields. Each one should earn its place.

What data will I end up holding? Every email and name you collect is your responsibility to protect.

Who holds my money, and for how long? Features that keep your money somewhere else come with a risk you should name.

Which features would I really miss? Not the ones that sound good, but the ones you would use this month.

What does each feature cost me? In fees, monthly subscriptions, setup time and buyers who give up.

Questions people ask

Isn't a tool with fewer features less professional? For buyers, professional usually means clear, fast and trustworthy. A checkout that shows the price, explains what to do and delivers the product immediately looks professional. Extra fields and pop-ups rarely help.

Can I still have a store? Yes. You can put payment links on any page you already have, or in a bio, a chat or a document. Many sellers do not need a separate store at all.

What if I want to collect emails? Collect them your own way, from buyers who choose to share them. The payment does not need to force it.

How do buyers without crypto pay? They first buy USDT or USDC through an exchange or a card service, which usually checks their ID once and charges a fee. MoonPay, for example, lists card fees of up to 4.5 percent, with a minimum fee of up to 3.99 dollars (MoonPay). After that first purchase, they can pay any seller who accepts digital dollars this way.

Will PayFi add these features later? The things described here were left out for reasons that have not changed. If PayFi ever changes its money rules, the change is written into its public ledger with a date.

Is it safe to pay without connecting a wallet? Yes. Sending money to an address is how most crypto payments work. It is connecting wallets and signing approvals on unfamiliar sites that carries the risk.

Back to Rina

A week after her lost weekend, Rina tries something smaller. She creates a payment link for one knitting pattern: wallet, name, price and the link to the PDF. It takes her about five minutes. She posts it under a photo of the finished scarf.

That evening, a buyer in Canada pays in USDC. The pattern unlocks on the buyer's screen. Rina sees 99 percent of the price arrive in her wallet. There is no account to set up, no email to confirm, and no app subscription renewing next month.

She keeps the store builder for now, but she cancels two of its apps.

A month later, she looks at what she actually used. Most of her sales came through payment links posted under photos and sent in chats. The store's mailing list has eleven subscribers. The discount-code app was used once. She lets the store subscription lapse and moves her patterns to a simple page with a link under each one.

Her buyers do not notice anything missing. They notice that paying takes a minute.

The old belief says the best tool is the one with the most features. The more useful belief is that every feature has a cost, paid in money, in time, in risk, and in buyers who give up halfway. The simplest way to get paid is the one that does exactly what you need and leaves the rest out on purpose.

Sources

• Pendo, "The 2019 Feature Adoption Report": pendo.io
• Pendo, "Pendo Data Suggests $29.5 Billion in Global Cloud R&D Investment Squandered When Software Features Go Unused": pendo.io
• Baymard Institute, "Cart Abandonment Rate Statistics": baymard.com
• CyberScoop, on IBM's Cost of a Data Breach Report 2025: cyberscoop.com
• MoonPay, "Pricing disclosure": moonpay.com
• Stripe, "Pricing": stripe.com/pricing
• BleepingComputer, "Cryptocurrency wallet drainers stole $494 million in 2024": bleepingcomputer.com
• PayFi home page, checkout page and public ledger: payfi.cx, payfi.cx/ledger