What a business should accept is set by what its customers already hold and how each coin behaves at the till, not by which coin is the better investment. This guide is for merchants choosing or switching a crypto provider. It gives you a decision framework built on four things a checkout actually sees — transfer cost, settlement speed, price behaviour, and how widely a coin is held — backed by on-chain measurements taken on 12 September 2026, plus a straight answer on how many coins to turn on.
The short path: start with what your customers hold, check each candidate against the four axes below, then enable the smallest set that covers your traffic. Everything after this is that path in detail.
A note on the numbers. On-chain costs move with network load. The figures here are a snapshot from public RPC endpoints on 12 September 2026, not fixed prices. The method matters more than the decimal, and the relationships between coins hold even when the absolute cost drifts.
Start with what your customers hold
The coin you accept is chosen by your audience, not by the market. A merchant gains nothing from a coin’s investment case: throughput, roadmap, or projected price do not reach the checkout. What reaches the checkout is whether the customer already holds the coin and can pay without a detour through an exchange. So the first question is demographic, not financial: what do the people who buy from you actually keep in their wallets?
For most merchants the answer starts with stablecoins. By issuance, dollar-pegged tokens dominate the market: total stablecoin issuance stood at $312.8 billion, with Tether (USDT) at 58.6% and USD Coin (USDC) at 23.8%, roughly 82% of all issued stablecoins between the two, as measured by DeFiLlama on 4 September 2026. That is issuance, not payment volume, but it tells you where liquidity sits. A buyer paying in crypto is very often paying in a dollar stablecoin, and the practical starting set for a Western or international merchant is USDT and USDC.
The second layer is the large held coins: Bitcoin and Ethereum. People hold them, some prefer to spend them, and both are supported almost everywhere. Past that, the tail is customer-specific. A gaming audience skews to fast, cheap networks; a privacy-minded buyer base raises demand for coins built around it. You learn this from your own order data faster than from any market ranking, and you can widen the set from the full list of supported coins once the demand is visible.
The four axes a checkout actually sees
A checkout judges a coin on four measurable axes: how much the transfer costs, how fast it settles, whether its price holds between payment and settlement, and how widely it is held. Investment metrics such as market cap, yield, and active addresses do not appear on this list, because none of them changes what happens when a customer pays. The table below reads every candidate coin against these four, using costs measured on-chain on 12 September 2026.
| Coin / rail | Transfer cost | Settlement speed | Price behaviour | How widely held |
|---|---|---|---|---|
| USDT on TRON (TRC-20) | ~$1.08–2.18 | under 30 sec | stable (pegged) | widest |
| USDT on Ethereum (ERC-20) | under $0.01 | 1–3 min | stable (pegged) | wide |
| USDT / USDC on Solana | under $0.01* | under 1 sec | stable (pegged) | growing |
| Bitcoin (BTC) | ~$0.11 at low congestion | 10–60 min | volatile | widest |
| Ethereum (native ETH) | ~$0.003 at current gas | 1–3 min | volatile | wide |
| Litecoin (LTC) | cents | minutes | volatile | medium |
| Solana (native SOL) | under $0.01 | under 1 sec | volatile | growing |
*Solana adds a one-time rent of about $0.15 the first time you receive a token if the receiving account does not yet exist. Subsequent transfers stay under a cent.
One line in that table contradicts the usual advice, and it matters enough to state on its own. On 12 September 2026, sending USDT over TRON (TRC-20) cost roughly $1.08 to $2.18, while the same USDT over Ethereum (ERC-20) cost under a cent. TRC-20 is repeated everywhere as the cheap default. Right now it is the expensive one, by more than a hundredfold, for a sender without staked energy on TRON. The reason is that TRON prices transfers in energy, and a plain USDT transfer to a fresh recipient burns enough of it to cost real money at the current energy price, whereas Ethereum gas has been sitting near its floor. This is exactly why a merchant measures the network fee rather than inheriting a rule of thumb. That rule of thumb held for years and does not hold today.
Read the axes together, not one at a time. A coin that settles in under a second but costs a dollar to send is fine for a $500 invoice and wasteful for a $3 tip; a coin that is nearly free but confirms in an hour suits a subscription renewal and frustrates a point-of-sale queue. The right coin is the one whose profile matches your average ticket and your customers’ wallets, which is why the next four sections take the pairings buyers actually ask about and answer each one from the till, not from a trading desk.
Ethereum or Solana for a merchant
For a merchant, this is a question of cost and speed, not of which ecosystem is the better bet. The investor comparison — which chain wins on throughput, developer count, or price outlook — has no bearing on a checkout. What the checkout sees is that stablecoin transfers on Solana settle in under a second for well under a cent, while native Ethereum settlement takes one to three minutes and its cost swings with gas.
Put plainly: for accepting dollar stablecoins, Solana is the faster and cheaper rail today. USDT and USDC on Solana confirmed in under a second at under a cent in the 12 September measurement, against ERC-20 USDT’s one-to-three-minute settlement. Solana’s own one-time account rent of about $0.15 applies once per new receiving token account, then disappears from the arithmetic.
Ethereum keeps two advantages that decide real cases. It is more widely held, so a share of your customers will hold USDT or USDC only as ERC-20 and cannot pay any other way. And ERC-20 is the settled default for business-to-business flows where the payer already runs on Ethereum. The merchant answer is therefore not “pick one.” It is: default new stablecoin acceptance to Solana for speed and cost, keep ERC-20 on for the customers and B2B counterparties who live there, and let the price behaviour of the underlying stablecoin — not the chain — carry the predictability.
Litecoin or Bitcoin
Both are volatile, so neither gives you price certainty at the till. The choice between them is cost, speed, and reach. This is also where the “Litecoin vs Bitcoin Cash” question lands, and the answer rhymes: among the volatile non-stablecoin coins, the lighter chains are cheaper and quicker, Bitcoin is more widely held.
Litecoin settles in minutes for cents. Bitcoin confirmed for about $0.11 in the 12 September measurement, at a moment of low mempool congestion. That cost climbs when the network is busy, and confirmation still runs 10 to 60 minutes regardless of fee. So on the two operational axes, Litecoin wins: faster, cheaper, less variable in cost.
Bitcoin wins the axis that often decides it: reach. It is the coin a first-time crypto payer is most likely to hold and most likely to trust, and that recognition converts. For a merchant the split is clean. If you want the broadest possible “we take Bitcoin” signal and your tickets tolerate a 10-to-60-minute confirmation, Bitcoin earns its place. If you are optimising the payment itself, its speed and cost per transaction, Litecoin is the better of the volatile pair. Neither removes the core drawback both share against a stablecoin: the amount can move between the moment your customer pays and the moment you settle.
Stablecoin or a volatile coin
For predictability at the checkout, a stablecoin beats any volatile coin, full stop. A dollar-pegged token is worth a dollar when the customer pays and a dollar when you settle. Bitcoin, Ethereum, and Litecoin can all move in the minutes between. This is the “USDC vs Bitcoin” and “Tether vs Bitcoin” question, and from the till it is not close: for matching what you charged to what you receive, the stablecoin is the instrument built for the job.
That is why acceptance and settlement are separate decisions. You can accept a volatile coin because a customer holds it, and still settle in a stablecoin so your books don’t inherit the swing. A gateway that auto-converts on receipt does exactly this: USDT or USDC lands on your balance whatever the customer paid with. On Speend, that auto-conversion into a stablecoin at receipt runs at 1% for most coins and 0% on Bitcoin, on top of a base fee from 0.5%. The specific rate depends on the coin, so confirm the pairing for the coins you take.
Want to accept crypto payments on your website?
Fast setup and KYC/KYB, fee starts from 0.5%
Contact UsThe trade-off to weigh is not about safety but about who bears the volatility. Accept-and-hold a volatile coin, and you are taking a position on its price whether you meant to or not. Accept-and-convert, and you pay a small conversion fee to hand that risk off and keep your revenue in dollars. For most merchants selling goods and services rather than speculating, converting to a stablecoin at receipt is the default that keeps a payment a payment.
Crypto or a card
Against a card, crypto’s merchant case is finality and cost; the card’s case is familiarity and buyer protection. A confirmed on-chain payment cannot be reversed by the payer — there are no chargebacks, which for a card merchant are both a fee line and a fraud-exposure line. The customer’s protection that chargebacks provide becomes the merchant’s cost, and crypto removes both sides of that.
On cost, the comparison is stark once you have the on-chain numbers. A card transaction typically carries a percentage-plus-fixed interchange and scheme fee measured in the low single-digit percent. A confirmed Bitcoin transfer cost about $0.11 in the 12 September measurement, and a stablecoin transfer on Solana under a cent — flat, not proportional to ticket size. On a large invoice the gap is decisive; on a small one, card economics can be competitive, and the honest read is that crypto’s cost advantage widens with ticket size.
Where the card still wins is onboarding and instant reversibility for the buyer. Everyone has a card, not everyone holds crypto, and the buyer-protection framework around cards is something a share of customers expect. The merchant conclusion is additive, not either-or: crypto is the stronger rail for finality, cross-border reach, and cost on larger tickets, and it sits alongside cards rather than replacing them for most businesses. Finality is the headline — you keep what you were paid.
How many coins to open
Open the smallest set that covers your customers, because every network you turn on has an operational price the “300 coins” figure hides. A gateway advertising 300+ coins and 18 networks is describing a shopfront, not the set you should run. The right number is decided by your order data, and it is usually small.
Each network you enable costs you three things beyond the transfer fee. It is an address issued per order that has to be generated and monitored. It is a reconciliation point, one more place where an incoming payment has to be matched to an order and reconciled in your books. And it is a failure scenario: an underpayment, an overpayment, a payment to an expired invoice, or a transfer stuck at low fee, each needing a defined response. Turn on ten networks and you own ten of each; turn on three and you own three.
A workable default for most merchants:
- Two stablecoins — USDT and USDC — as the core, since that is what most crypto payers hold.
- Two to three networks under them, chosen by cost and speed: Solana for fast, cheap settlement, Ethereum (ERC-20) for the customers and B2B flows that live there, and TRON only if your audience specifically uses it — measured, not assumed, given its current cost.
- One or two volatile coins if your customers hold them — usually Bitcoin for reach, optionally Litecoin for cheaper, faster settlement — with auto-conversion to a stablecoin on receipt.
That covers the wallets most buyers actually carry while keeping your reconciliation surface small. You can widen from the full list of all coins later, driven by real demand in your order data, and each addition is a deliberate choice with a known operational cost, not a checkbox turned on because it was there.
FAQ
Which coins should I start accepting?
Start with what your customers hold, which for most merchants means the dollar stablecoins USDT and USDC, then add Bitcoin and Ethereum for reach. By issuance, USDT and USDC together make up roughly 82% of all stablecoins (DeFiLlama, 4 September 2026), so a crypto payer is most often paying in one of them.
ETH or Solana for accepting payments?
For accepting stablecoins, Solana is the faster and cheaper rail: USDT and USDC settle in under a second for under a cent, versus one to three minutes on Ethereum with a gas-dependent cost. Keep Ethereum (ERC-20) enabled for customers and B2B counterparties who hold their stablecoins only there.
Litecoin or Bitcoin?
Litecoin settles in minutes for cents; Bitcoin is more widely held but confirms in 10 to 60 minutes and costs more when the network is busy. Both are volatile, so pick Bitcoin for reach and Litecoin for cheaper, faster payments — and convert either to a stablecoin on receipt if you want price certainty.
Stablecoin or a regular coin?
A stablecoin holds its value between the moment a customer pays and the moment you settle, and a volatile coin can move in that window. Accept a volatile coin if a customer holds it, but settle in a stablecoin: a gateway that auto-converts on receipt lets you do both.
How many coins should I open?
As few as cover your customers, usually two stablecoins over two or three networks plus one or two volatile coins. Every extra network is an address per order, a reconciliation point, and a set of failure cases to handle, so the “300 coins” figure is a shopfront, not a target.
Crypto or a card for a business?
Crypto settles with finality (no chargebacks) and costs a flat on-chain fee rather than a percentage, which wins on larger tickets and cross-border payments. Cards win on universal familiarity and built-in buyer protection, so most businesses run crypto alongside cards rather than instead of them.
Where to take this next
Pull your own order data before you enable anything. The set that fits you is written in what your customers already hold and the size of your average ticket, and no market ranking substitutes for that. Decide the coins, then decide the networks under them on cost and speed — re-measuring the on-chain fee on the day, since the TRC-20 example on this page shows how fast a “cheap default” can invert.
If you accept a volatile coin, decide up front whether you are holding it or converting it, because that single choice decides whether a sale stays a sale or becomes a position. For most merchants, convert on receipt and keep the payment in dollars. The rest is maintenance: a small, deliberate set of networks you can actually reconcile beats a long list you enabled because it was on offer.
