Bitcoin can reach a million merchants. Why aren’t people spending it?

Bitcoin payments have reached more than a million U.S. sellers through Square, yet the group behind much of the network’s payment software says everyday spending remains scarce. Spiral’s October 27 report is an unusually frank assessment from a Block-backed developer: seven years of easier Lightning tools have expanded the places Bitcoin could be used without giving most holders a reason to use it there.

The contrast is visible at the register. A seller can display a Bitcoin option and still take no Bitcoin payments all day. A customer who acquired the coin as a long-term holding may see little reason to spend it on lunch when a card draws dollars from an account already used for expenses. Square can remove the merchant’s exposure to Bitcoin’s price by converting receipts, but it cannot decide which money a customer is willing to part with.

In a reported titled The Long Road to Bitcoin Payments, Spiral calls the gap mainly a demand problem. Its conclusion deserves a close look because “enabled merchants” and “transactions completed” are often placed side by side as if one followed from the other. No publicly comparable series of payment volumes or repeat customer purchases accompanies the million-seller figure. Until it does, reach is a measure of distribution, not adoption at checkout.

A merchant count describes the option, not the sale

Square’s rollout has made Bitcoin acceptance available to roughly one million U.S. sellers, according to coverage of the rollout. Eligible merchants can take a Lightning payment and receive dollars by default, with the conversion handled in the payment flow. A seller who prefers not to accept Bitcoin can opt out. The arrangement is a significant reduction in the work an individual restaurant or shop would otherwise face: sourcing software, monitoring exchange rates and deciding how to manage a volatile receipt.

It also makes the count ambiguous for a reader. Being eligible and enabled is not the same as having a prominently displayed Bitcoin checkout, a trained cashier or a customer who has paid with the method. A million merchants each making no Bitcoin sale would still be a million enabled merchants. At the other extreme, a relatively small group of active shops might account for nearly all payments. Without the number and value of completed transactions by merchant cohort, neither conclusion can be established from Square’s footprint.

JUST IN: Toyota vehicles can now be purchased with Bitcoin in Bolivia

TOYOSA unveiled the new payment option at Expocruz 2026, with BitGo Bank & Trust providing the institutional-grade digital asset wallet infrastructure alongside payments in bolivianos, USD and USDT. pic.twitter.com/mlfEJ7b5Se

— crypto.news (@cryptodotnews) September 19, 2026

The distinction matters to Block’s own product case. An integrated option may save a seller the effort of installing a separate Bitcoin terminal. The marginal cost of making the option available can be small even if immediate demand is low. That can be a rational long-term distribution strategy without being evidence that Bitcoin has displaced cards or cash. Square and Spiral operate within the wider Block ecosystem, but the report’s ambition for Bitcoin as money should not be mistaken for an independently audited measure of what Square’s users have spent.

The merchant can receive dollars even when the customer pays Bitcoin. That solves a crucial accounting problem for a business whose rent and wages are denominated in dollars. It does not establish that the merchant has begun holding Bitcoin or that a Bitcoin economy has formed between customers and suppliers. A processor could settle the seller in cash and buy or sell Bitcoin in the background. The useful customer metric is the fraction of checkout payments actually funded with Bitcoin, not the number of locations capable of displaying the button.

A comparable account would show how many sellers had at least one Bitcoin sale in a specified month, how many had repeat sales, their total transaction value, and the share of all Square payments that those purchases represented. It would distinguish refunds and failed attempts from completed sales. Reporting an opt-in setting or a seller directory without those denominators cannot answer the demand question Spiral has raised. Nor should a lack of public figures be turned into an assertion that no one is using the feature.

Why a holder reaches for dollars

Bitcoin’s long-term investment story works against its short-term spending story. Someone who expects the coin to appreciate may prefer to pay with a currency they already hold for everyday bills. The preference is not universal, and a holder can replenish coins after spending them. Replenishment, however, requires another purchase and may bring exchange spreads, transfer steps and record keeping. A tap of a card rarely asks the buyer to consider the future value of the funds used for groceries.

For a merchant quoting a dollar price, Bitcoin volatility introduces an additional choice about conversion. Square’s default dollar settlement removes much of that exposure for its seller, which is an important product improvement. A person who directly receives and keeps Bitcoin still bears the market risk between sale and later expenses. A customer sees a dollar price translated into a BTC amount at checkout, and the exchange rate may have shifted since the coins were acquired. Stablecoins offer a dollar unit at both ends, although their issuers, reserves and redemption arrangements create different risks.

U.S. tax treatment adds another step. The IRS guidance on digital assets treats a purchase paid in a digital asset as a disposition that can generate a gain or loss. A person who bought Bitcoin for less than its value at the moment of spending may need to calculate the difference, even on a small purchase. The amount of tax, if any, depends on the person’s basis, holding period and wider circumstances; the record-keeping obligation can still make the experience less like spending dollars.

The comparison with a card is often framed only around the seller’s processing fee. A lower network charge does not automatically produce a lower total cost for the buyer. The customer may have paid to acquire the coin, could face a conversion spread when replenishing it and needs to account for a disposal. The processor and seller also bear costs to turn an incoming Bitcoin payment into dollars and handle refunds. A credible fee comparison would follow the same purchase from funding to settlement rather than comparing only a card merchant rate with a Lightning routing fee.

Spiral identifies U.S. tax rules as one obstacle and asks whether a change in policy could make small payments easier. A proposed de minimis exemption would need legislation, an effective date and qualifying limits; a wish for it is not a current exemption. For now, a consumer who values simple accounting may choose dollars even when both the wallet and merchant support Bitcoin. Payment design cannot erase that legal difference by changing a checkout screen.

JUST IN: Volkswagen Group Singapore now accepts Bitcoin and crypto as payment for cars and after-sales services pic.twitter.com/t9lQjVPtFk

— crypto.news (@cryptodotnews) April 13, 2026

Lightning has made Bitcoin payments easier to use

Lightning lets participants move Bitcoin through payment channels without putting every small payment directly in a Bitcoin block. The network has matured since early deployments, and wallets increasingly hide channel management and liquidity from users. Spiral’s report says custodial Bitcoin can now be integrated into an app with effort comparable to custodial stablecoins, while running self-custodial services at scale remains more difficult. Those claims speak to developers’ work, not to the number of people choosing BTC at a shop.

A better user experience can remove reasons a motivated customer gives up. Reusable payment addresses, clearer amounts, recovery methods and a single QR flow all matter if an app is already trying to make Bitcoin payments. Yet payment preferences are shaped by what the buyer receives and what the merchant asks for. A technically successful transfer does not tell us whether the same buyer returns next week. A wallet that handles a first transaction well is necessary for repeat commerce, but it is not sufficient to create it.

The transaction can be routed through Lightning while the economic payment remains denominated in dollars. Spiral distinguishes Bitcoin as a rail from Bitcoin as the money the customer chose to spend. If a wallet converts a customer’s dollar balance and settles a Lightning invoice, Bitcoin infrastructure carried value, but the customer did not necessarily decide to draw down Bitcoin savings. Both kinds of use can be important. Combining them in a single “Bitcoin payments” total would conceal the particular adoption problem under discussion.

A similar ambiguity appears when a shop receives dollars by default. The customer has spent Bitcoin, satisfying one measure of payment use; the merchant has chosen the conventional unit for its books. If the objective is to make Bitcoin a common medium of exchange among households, customer-side spending is directly relevant. If the objective is to build open settlement infrastructure that can carry different currencies, a dollar payment over Lightning also counts. Spiral’s report argues for both goals but says explicitly that they should be measured separately.

The strongest opposing view is that everyday spending is premature for an asset many people still treat as savings. Bitcoin could become less volatile as markets deepen, and users facing unstable local currencies might find self-custody more compelling than a shopper in the United States. Spiral acknowledges this argument rather than assuming every unspent coin signals failure. The evidence needed to compare the views is geographic and longitudinal: are payments growing where the advantages are greatest, and do users keep paying after an introductory incentive ends?

Stablecoins are already in the checkout flow

The early Shopify USDC pilot shows how an established commerce platform can add a dollar-token option within familiar merchant software. Availability did not itself disclose how many buyers chose it. Its relevance to Bitcoin is distribution: payment methods are increasingly built into tools sellers already use, so the competition is over both the customer’s preferred asset and the checkout interface where that preference is expressed.

Dollar-backed tokens solve the unit-of-account problem for a merchant selling goods priced in dollars. Providers including Stripe and Shopify have built stablecoin payment options, while PayPal operates its own dollar token. A stablecoin customer still needs a wallet, a way to acquire the token and a merchant willing to accept it. The practical difference is that an advertised ten-dollar item need not be repriced against a fluctuating Bitcoin balance between order and settlement.

Stablecoins do not make issuer risk disappear. A user relies on reserves, custody and redemption rules, and an issuer may restrict access under legal or operational requirements. Bitcoin’s self-custody model can be useful where an individual wants to transfer value without depending on a dollar-token issuer. The choice therefore varies by location and use. Spiral’s concern is more immediate: as familiar payment providers choose dollar tokens for new integrations, developers and merchants may build habits around those interfaces before a comparable Bitcoin option reaches their customers.

The sequence matters. A platform that already accepts cards and dollars can add a dollar-backed token without teaching the merchant a new pricing language. A Bitcoin payment option can be made equally easy for a processor to integrate, according to Spiral, but the customer must still prefer it at the point of sale. There is no inevitability that the first set of payment rails will win forever. Switching costs become more important, though, once wallets, merchant reports, refunds and developer tools are organized around one accepted asset.

Some stablecoin activity is cross-border treasury movement rather than retail shopping, so a total transfer-volume comparison can mislead. The right comparison depends on the question: completed merchant purchases in the same region, payer-to-payee remittances of the same size, or payments for an API call under the same protocol. Onchain transfers can include exchange rebalancing or movement between addresses controlled by one organization. A headline saying stablecoins exceed Bitcoin in all “payments” needs to separate these categories before it can describe customer preference.

Even when a stablecoin wins the checkout, Bitcoin might remain the customer’s savings asset. A user can sell a small portion of BTC to acquire dollars or a stablecoin before spending. The purchase is then economically funded by Bitcoin wealth but transacted in another currency. Whether such a pattern is a loss for Bitcoin as a payment system depends on the goal being measured. It certainly does not show a shopper paying directly in BTC, the use Spiral wants payment firms to make easier and more attractive.

AI agents are beginning to pay with Bitcoin

Software agents can request an API, receive a price and pay automatically for a small amount of data or computing. Such purchases need predictable costs and a machine-readable payment flow. The x402 standard uses the web’s Payment Required response to help applications do this. Spiral points to early agent payments as a place where dollar tokens have established a lead, citing a roughly 99% USDC share in a particular x402 measurement. The denominator is x402 activity under that measurement, not all AI payments worldwide.

Block’s September Lightning contribution to x402 adds a Bitcoin route to a standard that was already processing stablecoin requests. The relevant technical addition was visible in the project’s repository before Block announced it. Support in an open protocol means a developer can implement the option. It does not mean Square, Cash App or every agent platform has begun using Lightning for x402 transactions. Block had not published a rollout date or payment share for its consumer products.

A comparison between token and Bitcoin options should use the same period and method. Count successful requests, value settled, failed attempts, refunds if relevant and repeat service buyers. If a provider reports millions of x402 transactions across all networks, those are not millions of Lightning payments. The network identifier in the payment proof can distinguish routes, but public dashboards must expose a comparable sample. A service could also choose a Lightning invoice for some small requests while preferring dollars for predictable accounting elsewhere.

Lightning Labs has offered another route. Its Wavelength mainnet opening lets developers give agents a Bitcoin wallet without each running a Lightning node and managing channels. Its initial account balance limit is 300,000 satoshis, and its published service fee on Lightning sends and receives is 0.25% plus routing charges. The service is in alpha, while stablecoin support remains planned. These terms describe an accessible experiment, not a demonstrated market share against USDC.

An agent can pay for an API without a human seeing the transaction. The question is who funds its wallet, which asset the seller wants and what controls prevent unbounded spending. A user who deposits Bitcoin into an agent wallet has chosen BTC as working capital; another might top it up with dollars if a future interface performs conversion. Agent commerce could produce many small payments while very few humans change their own checkout behaviour. It is a useful test of the rail and perhaps of the currency, provided the two are measured distinctly.

Incentives can buy a trial, but not a habit

Spiral asks developers to propose reasons existing holders would spend, including discounts and rewards. A rebate can overcome a customer’s reluctance to take an unfamiliar payment route for the first time. Its value as evidence appears after the promotion ends: does the same buyer return when a card, stablecoin and Bitcoin all compete on ordinary terms? A campaign reporting only the first month’s transaction count could reflect people collecting a subsidy rather than discovering a lasting preference.

A merchant incentive has a different effect. If a seller receives a fee benefit or faster access to proceeds, it may present the Bitcoin option more prominently and train staff to handle it. The customer still must have a compatible wallet and an answer to the question of why BTC is the easiest money to spend. A merchant could choose to pass some savings to the buyer, changing that calculation. Whether the combined economics work depends on the full cost of acquiring, transmitting, converting and accounting for the payment, not the nominal Lightning fee alone.

Micropayments could offer a more distinctive use than replicating a card purchase. Paying a creator by the minute, buying one data request or compensating a remote computer for a tiny unit of work may be awkward under card minimums or subscription billing. Bitcoin’s divisibility and Lightning’s small transfers can make such experiments feasible. A product also needs useful content, reliable delivery and protections against accidental repeated charges. Payment infrastructure solves the money movement; the service must earn a reason to pay.

Spiral also points toward places where capital controls or an unstable local currency make self-custody valuable. A dollar stablecoin may still be attractive there because its nominal price is steadier, though issuer restrictions and access can matter. The evidence for Bitcoin’s advantage would come from people choosing to hold and spend it repeatedly in a particular market, not from assuming all people under inflation have identical needs. Remittance, local merchant spending and saving should be counted separately. A transfer into savings does not automatically return to circulation as a checkout payment.

The open call for a five-year “moonshot” is a request for proposals and evidence, not an announced product with a budget, completion date or guaranteed payment volume. Spiral invites payment providers to share data. Its willingness to say usage is weak should make those data central to evaluating any project it supports. A fresh wallet feature may be genuinely useful and still fail the question if it increases installation counts without completed, repeat spending.

How to tell whether the million merchants are being used

Start with the time period and the denominator. A monthly count of active Bitcoin-accepting sellers divided by all eligible sellers would say more than the overall million. The same report should show how many transactions each active seller processed and the share of checkout value they represented. Merchants who made one test sale should be distinguishable from those who see routine purchases. The difference determines whether acceptance has become a working option in commerce or remains a feature waiting for demand.

Next follow the buyer. How many distinct customers paid twice or more within 30 or 90 days, and how many continued when a discount ended? Privacy requirements may prevent a processor from publishing personally identifiable records, but aggregate cohorts can answer the question without naming individuals. A venue that advertises growth in payment volume should say whether one large merchant, a new seller cohort or repeated everyday payments drove it. Otherwise a single integration can make a network look broadly popular when most sellers saw no change.

Finally state the asset at each end. A customer paying BTC while the seller receives dollars is Bitcoin spending by the customer and a dollar receipt by the merchant. A dollar payment carried over Lightning is a use of Bitcoin-linked infrastructure, with no necessary customer decision to spend BTC. A stablecoin transfer between two exchanges is neither a merchant purchase nor proof of a consumer choosing dollar tokens at checkout. Separating the flows gives the million-merchants claim a fair test instead of making the categories compete on incompatible totals.

One simple calculation can illustrate the missing denominator without pretending to know the answer. If 10,000 of a million enabled merchants took one Bitcoin payment in a month, the active share would be 1%. If 100,000 did, it would be 10%. Both scenarios are consistent with a million enabled sellers, and neither is asserted as Square’s actual performance. Publishing the active seller count would replace speculation with an observed ratio. Publishing repeat buyers and value would show whether that ratio reflects meaningful consumer spending.

The report’s competing interpretations can also be tested over time. If Bitcoin’s volatility declines and wallets become simpler but the active buyer share stays flat, the demand problem would look more persistent. If repeat spending grows in markets with weak local currencies and later spreads to other regions, the patient store-of-value argument gains evidence. If an incentive produces one-off purchases that vanish when the offer ends, it has not solved habitual use. The useful result may arrive gradually rather than in a single launch-day announcement.

The five-year question begins with this year’s receipts

Congress has considered limited digital-asset tax changes, including a House proposal addressing small network fees. Its $10 fee provision is not a general exemption for a coffee purchase funded in Bitcoin, and a committee proposal is not law. An actual payment exemption would need operative language defining the qualifying transaction, a threshold and an effective date. Until those steps occur, merchants can simplify their side of checkout while consumers still face the existing disposition rules.

Spiral’s appeal is for a credible path to Bitcoin payment adoption within five years. It can support experiments, and Block can put payment options in front of a large seller base. The first measurable question is more modest: how many of the sellers and customers who already have access actually use it? The public report establishes availability and acknowledges weak demand, but does not supply a comparable Square transaction series.

A future release of merchant cohorts, repeat-customer rates and payments actually funded in BTC would let readers judge progress. The figures would also expose uncomfortable but useful findings, such as a product used heavily by a few shops or a high trial rate followed by little return traffic. Those outcomes can guide product changes. A payment route is valuable only to the extent someone chooses it and the recipient can use what arrives.

The available evidence already rules out one easy explanation: people cannot spend Bitcoin merely because no shops can accept it. Square has reduced that distribution barrier for a million sellers. The remaining work involves why buyers would choose BTC, whether the complete transaction is convenient and what a merchant does after accepting it. Spiral’s call is most useful if payment providers respond with the data needed to tell those answers apart.

What to watch

• Active Square sellers: Seek a dated count of merchants with completed BTC sales, divided by the number enabled, plus repeat seller activity.
• Returning buyers: Compare first-time and repeat Bitcoin payers after any discounts or rewards expire.
• Asset at checkout: Distinguish customer-funded BTC purchases, dollar payments over Lightning and merchant settlement in dollars.
• Agent transactions: Look for x402 and Wavelength volumes broken out by asset, network, successful requests and fees.
• Tax treatment: Track any enacted U.S. rule for small Bitcoin purchases and its effective date rather than treating proposals as current law.

FAQ

Disclaimer: This article is for information and educational purposes only and does not constitute financial or investment advice. Figures reflect regulatory filings and reporting available at the time of writing and change with each disclosure. Nothing here is a recommendation to buy, sell, or hold any security or asset. Always do your own research. Information is accurate as of October 28, 2026.