State of Crypto Payments 2026: Statistics That Matter to Merchants

What do these crypto payment statistics count?

They count different things, so we keep the categories apart. Our evidence cutoff is 11 October 2026, which isn't the end date of every dataset. The newest checkout data is CoinGate's January–June 2026 report. Consumer evidence runs to the Federal Reserve's October 2025 survey, and the ECB's comparable study is labelled 2024.

Here, a merchant crypto payment means a customer has completed a payment for goods or services through a crypto checkout or invoice. A provider may convert the incoming asset before paying you, and that still counts as crypto acceptance. A later wallet withdrawal, supplier payout or token swap is a different event.

A stablecoin can show up in all of those flows, so the token's name doesn't tell you whether anyone bought something.

Allium's own documentation separates raw transfers, supply events, balance snapshots and adjusted volume metrics, and we read headlines with those distinctions in mind.^[15]

Which measured findings are useful in 2026?

The useful findings name a population, a unit and a period. The table compiles published evidence and isn't a new survey. Provider rows are the providers' own reports, and the survey and blockchain rows carry their own limits.

There's no total at the bottom: adding these rows would mix dollars, people, orders and event logs, and could count the same money twice.

To choose a service once you've read the evidence, use our crypto payment gateway comparison. This report measures activity, not provider fit.

What does completed-order evidence tell a shop owner?

It shows people successfully paying through one processor, but not how big crypto is next to every other payment option in those shops.

CoinGate's annual report defines completed orders as successful shopper-to-merchant payments. Its total includes orders recorded outside the blockchain, called off-chain orders, while its asset and blockchain analyses leave them out.^[1] The H1 update is less clear about what each breakdown includes, so we haven't built a worldwide coin-share chart from these figures.^[2]

Growth is tricky to read. More completed orders could mean existing merchants got more orders, new merchants joined, the mix of businesses changed or the product changed. Unless the same merchants are tracked over the same period, you can't pull customer demand out of those effects.

You need your own totals to compare against. Count crypto orders alongside all completed orders, and payment attempts alongside successful payments. Revenue is a third question: a few expensive invoices can make up a lot of revenue and very little of the order count.

Our CoinGate review covers that service's practical terms. Its published activity is evidence about that platform, so ask how closely its merchant mix resembles yours.

Do ownership and merchant surveys prove mass adoption?

No. They give clues about demand within their own samples, and each asks different questions of a different population.

The Federal Reserve survey recruits people by random sampling and weights the results to represent US adults. Payments are a different question from investment and transfers to friends or family, and respondents can pick several uses, so the same person can appear in more than one percentage. Someone who used crypto once during the recall period counts as a user, just like a frequent payer.^[5]

The ECB also splits ownership from the reason for holding crypto, and warns that the ownership question changed between survey waves, which may affect comparisons over time.^[6] SPACE combines a questionnaire with payment diaries, and its methodology uses telephone and online interviews alongside separate German and Dutch studies.^[7]

The NCA survey ran online for selected sectors and job roles. Its public summary gives fieldwork and sample details, but the full weighting variables and subgroup sizes are available only on request. We don't stretch its acceptance result to every US sole trader, local service business or shop.^[8]

These findings can all be true at once, because ownership, occasional payment use and acceptance among decision makers aren't estimates of the same thing. A shop can enable crypto and get little spending, and an owner can hold tokens and still prefer cards.

Before paying for an integration, ask your existing customers which asset and network they'd use. Specific requests tell you more than whether crypto sounds interesting, and our processor selection guide helps turn them into a shortlist.

Why aren't stablecoin transfer totals merchant-sales totals?

Because the same tokens also move through trading, lending, treasury transfers and payment infrastructure, and not every public transfer shows what it was for.

The BIS study rebuilds transactions and their internal events instead of assuming each transfer is a purchase. It covers only the named stablecoins on Ethereum in the year it names, so it doesn't count worldwide merchant spending or show that every simple transfer is a retail payment.

Visa and Allium filter out repeated movements, high-frequency activity and other distortions. Their current methodology sorts activity into use-case categories, and even the payments category includes payroll and person-to-person transfers alongside goods and services. The retail-sized filter covers adjusted transactions below $250 and doesn't identify a merchant.^[12]

Visa's discussion of stablecoin velocity says transfer data alone can't reliably identify every on-chain purchase, and its earlier dashboard explainer also splits adjusted activity from small-transfer activity. Both are useful views of the network, but neither is a ledger of completed invoices.^[13][14]

Studies built on payment firms get closer to what a transfer was for. Artemis's May methodology gathers contributor data and estimates, acknowledges possible duplication where firms serve each other and describes incomplete sector coverage. Its later update discusses business payments and card-related use, which are broader than direct merchant checkout. If the group of companies changes between studies, the studies aren't comparing the same businesses over time.^[16][17]

For your business, the record that counts is the paid invoice matched to the sale. A blockchain transfer reference helps you match the money to your books but can't replace the order record.

What do stablecoin shares and crypto-linked cards mean for checkout?

They show several ways crypto reaches commerce, and each means different work for you. A direct wallet payment, a crypto-funded card purchase and a processor's later settlement are three different things.

BitPay's stablecoin figure is a share of payment value. The report also covers wallet buying, selling and swapping and has a payouts section, and we leave those out of merchant checkout. Its announcement is dated 22 December 2025, before the year ended, and doesn't give the exact data cutoff, so we keep the source's 2025 label with that caveat.^[4]

Visa describes cards that check a customer's stablecoin balance, reserve funds and convert as needed, while the merchant still sees a Visa card payment. That lets a customer spend crypto holdings through your existing card checkout without your shop having any wallet-payment integration.^[11]

Choose direct acceptance when it meets a customer need your current payment options don't, and pick the settlement currency as a second decision. For stablecoin checkout, check which networks are supported. Our BitPay review covers that provider's acceptance and settlement details.

Does crypto acceptance make bank access or settlement automatic?

No. Acceptance, conversion and spendable bank funds are three checkpoints, and a completed payment only shows the checkout worked. Whether you can pay a supplier from your bank today is a further question.

CoinGate's settlement finding shows the gap: customers can pay in crypto while merchants choose regular money, called fiat, for settlement. That choice says nothing about how long bank credit takes or what's deducted from your account.

BitPay's developer documentation separates automatic business-day settlement from the bank arrival that follows, and notes currency minimums. Those are different clocks, even within one product. We cite it to explain the process, not to promise that a new applicant would qualify or when money would arrive.^[19]

A dollar stablecoin makes it simpler to hold a balance in one unit, but you still have currency exposure if your bills are in euros or another currency. Circle publishes reserve disclosures and assurance reports, which describe backing, not your processor's payout performance.^[20]

Circle's non-EEA USDC terms, updated 12 December 2025, make direct redemption depend on an eligible Circle Mint account and other conditions, so holding the token doesn't give your business a direct bank-redemption service by itself.^[21]

Our BitPay fee breakdown shows why the headline rate alone isn't the whole bill.

What should a business measure before expanding crypto payments?

Measure completed sales, net proceeds and the work it takes to receive them. Global statistics can justify a look, but your own payment records should justify expanding.

Start with customers and invoices you already know. A small group of business customers may give clearer feedback than switching on every coin across the whole storefront. Set a review date and define success before launch: reliable fulfilment, an acceptable total cost, usable settlement and support work you can handle.

Count the provider's fee, network costs, quote or conversion margin and bank charges in the cost review, plus any account or one-off charges. An unknown charge isn't zero. Track refunded sales on their own and count the handling work as well as the money deducted. This report shows no universal saving because the account-specific inputs aren't available.

1. 
   01

   Define the payment you need

   Name the customer group, invoice currency, assets, networks and country. Decide what currency you need for expenses.

2. 
   02

   Check the whole payment path

   Look at merchant approval, the contract entity, accepted assets, conversion, payout conditions and your bank's requirements.

3. 
   03

   Measure completed orders

   Record attempts, successes, sales value, refunds and payment-method switches. Match each payment to an invoice.

4. 
   04

   Review usable proceeds

   Compare net funds, bank-credit timing and support work with your target before expanding the customer group.

If you'd rather outsource, our NOWPayments vs CoinGate comparison ties product differences to operating needs. If your previous checkout was Commerce, the Coinbase Commerce alternatives guide covers moving off it.

Self-hosting adds another measurement gap. BTCPay Server describes itself as software through which you become your own processor, so that activity may never appear in a hosted provider's order report. Hosted-provider totals can therefore miss it, which doesn't prove self-hosted use is absent.^[24]

For regulated or restricted businesses, approval is its own question, apart from demand. ESMA's 8 October 2026 opinion says MiCA-authorised providers should stop services involving non-compliant stablecoins for EU clients, with limited activity allowed to deal with existing holdings. So the provider, the asset and your customers' location all matter when you check availability.^[22] The FSB's 16 October 2025 review, based on August information, found inconsistent implementation across jurisdictions, and it isn't a current permission list for your business.^[23]

Rules and permissions depend on your country, your provider's legal entity and your activity. Our guide to high-risk payment options helps frame questions about provider policies. This is general information, not legal advice.

What are this report's limitations, and can you reproduce it?

You can reproduce the published comparisons but not every provider's private ledger. The sources list the publications we opened, and the findings table keeps each unit, data period and coverage.

We chose primary processor reports, original survey publications, dataset documentation and official research, read each source's published methods and kept figures whose coverage we could describe. We didn't run a merchant survey, open accounts, make payments or fill in missing market totals. The hypothetical examples use made-up inputs and our own arithmetic.

Geography is another gap. An IMF working paper estimates wallet regions from what's visible on the blockchain rather than from customer invoices, so it can't tell you the destination country of every merchant sale.^[18]

Provider coverage, product changes, anonymous wallets, rounded survey answers and unclear descriptions of what totals include all rule out a worldwide estimate. We don't add up overlapping datasets, forecast year-end totals or treat one month's annualised pace as a measured year. A change in who contributes data can also look like growth.

To update this report, repeat the same definitions first. Record the new period, group of respondents or companies, and exclusions before replacing a number. If a researcher changes its method, keep the earlier and later results labelled apart until you can confirm they're comparable.

Crypto payment statistics FAQ

Change log: 11 October 2026, first published, with each figure's data period kept distinct from its publication date.