Crypto Payment Gateway Dev Cost for Casinos in 2025

Your P&L takes three hits here, not one.

The integration bill lands once, and then it's done. The volume fee lands every month you stay open, either priced per transaction or tucked inside an FX spread where you can't easily see it.

Then comes the third hit, the one nobody invoices you for: the ops, support and compliance work that sits on your desk regardless of what the vendor's deck implies. Build the thing in-house and yes, the vendor fee disappears.

What shows up instead is payroll. Engineers, node infrastructure on every chain you touch, and a maintenance job with no finish date. Add up year one honestly and building almost always costs more.

Why is crypto payment gateway pricing for iGaming so hard to compare?

Because the quotes sitting on your desk describe four different products.

Vendor one takes a percentage of deposit volume. Vendor two wants a monthly licence and a thinner percentage on top. Vendor three shouts "zero fees" and makes its money on the spread when it turns your BTC into euros.

Vendor four wraps hosted cashier, KYC orchestration and settlement into a single number, while the cheapest quote in the stack quietly assumes you already own all three.

Underneath the pricing models sit the variables that genuinely move your cost base.

Deposit count versus deposit value

Blockchains charge per transaction. They don't care about your euros. Fifty thousand deposits at €30 apiece carries far more network cost than five thousand at €300, even though the volume line looks broadly the same.

If a vendor quotes a rate without asking your average deposit size, they haven't priced your business. They've priced an imaginary one.

Coin mix

Bitcoin on-chain, Bitcoin over Lightning, ETH, Tron USDT, Solana stablecoins. Different fee profiles, different confirmation behaviour, different ways of going wrong.

A book settling mostly in TRC-20 USDT lives in a different cost universe from one where players fire BTC on-chain in the middle of a fee spike.

Jurisdiction and licence load

Curaçao and Malta are not the same compliance job. Neither is the Isle of Man. That gap shows up in your AML documentation, your transaction monitoring expectations and your headcount plan. It does not show up in the transaction fee.

Payout profile

Promise ten-minute withdrawals and you need working capital sitting in hot wallets, plus a treasury process to manage it. Real money, real risk. Never on a pricing sheet.

So normalise before you compare. One number, one basis: fully loaded cost per successful deposit at your actual volume, your average deposit size and your coin mix.

What are the actual line items in a first-year crypto payments budget?

Here's the shape of it. Every figure below is an order of magnitude, driven by your volume, licence and coin mix. It isn't a price list. And any vendor who hands you a price list before asking you a single question is guessing.

Integration and go-live

On a white label deployment this is usually the smallest number in the whole budget. It's also the number operators spend 80% of their negotiating energy on.

A standard integration into an established platform tends to land in the low-to-mid five figures. If the vendor already maintains a connector for your platform provider, it can be a fraction of that.

The number climbs when you want the cashier flow rebuilt from scratch, non-standard KYC orchestration, or a hookup into a proprietary player account system rather than a mainstream platform.

It climbs hard when your platform's payment abstraction layer is badly documented, or owned by a third party who bills for every change request. That second invoice comes from your platform provider, not your payments vendor, and it gets left out of budgets with depressing regularity.

Ask early. Ask in writing.

Four one-off items sit underneath the headline integration figure. They rarely get costed properly.

Sandbox testing and UAT

Somebody has to run deposits through every coin and every chain you plan to support, then deliberately break them: underpayment, overpayment, wrong network, deposit sent after the address expired, withdrawal cancelled mid-flight, two deposits to the same address in ninety seconds.

Budget real QA days here, not an afternoon between other jobs. A structured week in sandbox catches the edge cases that would otherwise land as live tickets on your first busy Saturday night.

Reconciliation setup and ledger mapping

Your finance team needs crypto deposits arriving in the ledger in a shape their existing chart of accounts recognises: gross deposit, network fee, vendor fee, conversion rate applied, net credited to the player.

Getting that mapping right is a one-time project involving your controller, your platform's reporting layer and the vendor's export format. Skip it now and you'll rebuild it in month three, under pressure, while someone from finance stands behind your chair.

Staff training

Support agents need to understand what "0/1 confirmations" means and why a player's BTC hasn't landed yet. Finance needs to know why the euro figure moved between deposit and settlement.

Compliance needs to read a blockchain analytics risk score without escalating it to the board. Two or three sessions plus written runbooks. Cheap. Almost always skipped.

Custom branding and wallet configuration

Optional, and priced separately by any serious vendor. Slapping your logo, palette and copy on a hosted cashier is usually a small fixed fee.

Deciding which coins appear, in what order, with which minimums, on which chains, per market, is genuine work that moves conversion.

Get it quoted as a line item instead of assuming it's free because someone called it "just config."

Where operators overspend

The pattern I see over and over: the operator commissions a bespoke crypto cashier before a single player has deposited a satoshi with them.

Six weeks of design and front-end work. A custom coin selector. Animated confirmation states. A branded QR experience with a little pulse effect. All of it built on assumptions about how their players behave in crypto that nobody has tested, because crypto isn't live yet.

Then it ships, and the data says something inconvenient. Ninety percent of deposits come in as USDT on Tron, not the four coins that ate the design budget. Players drop out at the network selection step, not the screen you polished for a fortnight. The average deposit is half what marketing forecast.

Better sequence: launch on the vendor's standard hosted cashier. Collect eight to twelve weeks of real deposit data. Then spend on customisation, aimed squarely at whatever the funnel is actually telling you. Same budget, two or three times the return, because it's pointed at a measured problem instead of an invented one.

Second overspend: launching with nine chains because nine looks thorough. Every extra chain drags in testing, reconciliation logic, support scripts and treasury handling. Most operators find three or four assets carry north of 90% of volume. Add the rest when players start asking.

Recurring transaction or spread-based fees

Once volume ramps, this line eats your first-year total.

Bitcoin payment gateway fees for casinos generally sit between sub-1% and the low single digits. The bottom of that band goes to high-volume operators pushing stablecoin-heavy flow. The top goes to low-volume brands, exotic coins, or deals where the vendor is also carrying FX and settlement risk for you.

Pay attention to how the fee gets taken. A flat percentage of deposit value behaves nothing like a percentage plus a per-transaction floor. Say that floor is €0.50 and your average deposit is €25. The floor alone is 2%, before the percentage component has done anything at all.

Spread pricing deserves harder scrutiny than anything else on the sheet. If the vendor converts crypto to fiat and pockets the difference, the advertised fee can honestly be zero while the effective cost runs anywhere from thirty basis points to comfortably over 1%.

Ask for the spread as a number. Ask which reference rate it's benchmarked against. Ask to see it on a real settlement report from a real client. A vendor who won't disclose the spread has just told you roughly what the spread is.

Network and on-chain fees

Someone pays the miner or the validator. Sometimes it's the player. Sometimes it's you. Sometimes it's folded into the vendor's fee. Find out which. In writing.

The exposure is lopsided, and deposit size drives it. When the Bitcoin mempool backs up, on-chain fees climb past the point where a small deposit makes any sense whatsoever.

A €20 deposit carrying a €6 network fee isn't a product. It's a complaint with a timestamp. Operators with low average deposits carry this risk far more heavily than high-roller books, and they carry it precisely when volume peaks.

Payouts, treasury and conversion

Deposits get all the attention in vendor calls. Payouts quietly cost more per unit.

Four separate costs hide in here:

Withdrawal network fees. Every payout is an on-chain transaction unless you're on Lightning or an internal ledger. In most casino configurations you're paying that fee, not the player, because charging a winner to collect their winnings is a support ticket you don't need. Batching helps. Chain choice helps more.

Conversion in and out of fiat. Euro-denominated balance sheet, payouts in BTC? You're converting twice: crypto to fiat on the way in, fiat to crypto on the way out. Two spreads, both yours. Some operators cut this by holding a working float in the coins players actually use, which swaps spread cost for price exposure. That's a treasury policy decision, not a payments one, and it belongs in a written policy before launch rather than improvised on a volatile Tuesday.

Hot wallet float. Promise fast withdrawals and funds have to sit ready. That capital earns nothing and carries custody risk. The faster the promise, the fatter the float, the more the promise costs you.

Approval workflow labour. Somebody reviews the big payouts. Somebody signs. Somebody investigates the ones that smell wrong. At volume that's a rota, not a favour, and you need at least two trained people so a single holiday doesn't stall withdrawals.

Model this properly. A withdrawal-heavy VIP book can spend more on payout processing and treasury conversion than on deposit fees. Most of them discover it in month four, in a board meeting.

Compliance, AML and monitoring

A white label doesn't make this go away.

Your vendor may hand you blockchain analytics and sanctions screening. Great. Interpreting the alerts, filing the reports and defending your decisions to a regulator stays with you, permanently.

Budget a slice of an existing compliance analyst's time, moving to a dedicated hire once volume justifies it, plus licences for whatever monitoring tooling your jurisdiction expects on top of the vendor's stack.

For a mid-size operator that's realistically a five-figure annual internal cost, even with a fully-featured vendor doing the heavy lifting.

Finance reporting and audit for a new asset class

This one blindsides people, every time.

Crypto is a new asset class on your books, and your finance function has to treat it like one. That means a valuation policy: which rate, from which source, at which timestamp, applied to deposits, withdrawals and anything you hold overnight.

It means a documented process for realised and unrealised FX movement on crypto sitting on the balance sheet. It means month-end close now includes reconciling wallet balances against on-chain reality, not just ticking off a PSP statement.

Then your auditor turns up with questions. Depending on how comfortable they are with digital assets, expect higher audit fees, a longer sample-testing cycle, and requests for evidence you've never had to produce: proof of wallet control, key management documentation, third-party custody confirmations, the complete audit trail from address generation to player credit.

Two practical consequences.

First: get your controller and your auditor in the room before you sign the vendor contract, not after.

Second: weight "clean, exportable, timestamped reporting" heavily when you compare vendors.

A gateway with genuinely good reporting saves your finance team days every month and saves you audit fees every year. That's worth several basis points on the transaction fee. Nobody puts it in the comparison spreadsheet.

Internal operations and reconciliation

The unglamorous line finance always underestimates.

Every deposit that lands mid-confirmation. Every player who sends the wrong asset to the wrong chain. Every withdrawal that needs human eyes. Each one spawns a ticket and a reconciliation entry.

Automation quality decides the whole line. A gateway that auto-credits confirmed deposits, handles wrong-chain sends gracefully and spits out clean ledger exports might cost you a fraction of an FTE. One that needs manual matching will cost you a full support head at volume.

And it'll be your best agent, because they're the only person who understands it. Then they leave.

Failed and stuck deposits: the forgotten support line

Give this line a name in your budget. Nobody does, and it's one of the most reliably expensive parts of running crypto.

A share of every crypto deposit attempt goes wrong in ways card processing simply never manages. The player underpays by the network fee and the amount doesn't match.

They send USDT on the wrong network. They copy an address that expired eleven minutes ago. Their transaction sits unconfirmed for two hours because they set the fee too low. They close the browser mid-flow and swear the money left their wallet, which it did, to a chain you don't support.

Every one of those is a live chat with an anxious human who believes you've taken their money. Average handling time is high, because the agent has to pull up an explorer, confirm the network, escalate to whoever can trigger a manual credit or a recovery, then circle back.

Rough shape. Say 3% of deposit attempts need agent intervention, and each one burns fifteen minutes of skilled time end to end. At 10,000 monthly deposits, that's 75 hours a month.

Half an FTE, and not a junior one. Same volume, but with a gateway that auto-credits underpayments and catches wrong-network sends before broadcast? You might be looking at 0.5% and five hours.

That gap is worth more money than the fee difference between most vendors on your shortlist. So ask every vendor three things: deposit failure rate, auto-resolution rate, and exactly what happens to a wrong-chain send. Then ask them to put the numbers in the contract.

Cost per successful deposit: the only number that compares

Here's the formula. Run it on every quote.

Cost per successful deposit = total first-year cost ÷ completed deposits in year one

Total first-year cost means all of it: integration, sandbox and training, branding and wallet config, vendor fees, network fees, payout and treasury conversion, compliance time, finance and audit uplift, ops and reconciliation hours, failed-deposit support hours, and any engineering time you burned here instead of on revenue features.

Completed deposits means deposits that actually hit a player's balance. Not attempts. Not initiated sessions. Landed.

Which brings up the thing nearly every cost model gets wrong.

Conversion rate belongs in your cost model, not just your marketing dashboard. It lives in the denominator. Two gateways, both charging 1.5%, both handling 10,000 monthly deposit attempts:

• Gateway A converts 78% of attempts. 7,800 successful deposits.

• Gateway B converts 91%. 9,100 successful deposits.

Identical fee. But Gateway B spreads the same fixed costs across 17% more successful deposits, and hands you 1,300 extra funded players a month who then go and play your slots. Gateway A is more expensive per successful deposit and dramatically more expensive per deposit attempt. The rate cards are twins.

What actually moves conversion: how many taps it takes to reach a payment address, whether the coin and network the player already holds is on the list, whether the QR scans first time on a four-year-old Android, whether underpayments auto-credit instead of failing, whether the confirmation screen tells a nervous player what's happening and when. Product decisions, all of them. Financial decisions, all of them.

So put conversion rate in the comparison sheet. If a vendor can't tell you their deposit conversion rate for casinos in your market with your coin mix, they either don't measure it or don't fancy the answer.

Three questions to force into every vendor quote

Ask these three, in writing, before you compare anything. The answers will tell you more than the rate card ever could.

1. "At my volume, my average deposit size and my coin mix, what's my fully loaded cost per successful deposit, and what assumptions are you using?" Not a percentage. A number, with the working shown. A vendor who has priced iGaming deposit flow before will turn this around in a couple of days. One who hasn't will resend the rate card with the font enlarged.

2. "What percentage of deposit attempts fail or need manual intervention, how many resolve automatically, and what exactly happens when a player sends the wrong asset on the wrong chain?" One question, two answers: your support cost and your conversion rate. Push for real figures from comparable clients, not a tour of the happy path.

3. "Show me the full cost of a €50 deposit and a €50 withdrawal, line by line, including network fees, spread, and who pays what." Two concrete transactions, start to finish. Deposit fee, network fee, conversion rate applied against a named benchmark, withdrawal fee, payout network fee. This is where "zero fees" quietly becomes 1.4%, and where you learn whether you or the player eats the on-chain cost.

A bonus fourth, if you want to be thorough: ask what happens to your deposits during a Bitcoin fee spike. Get that one in writing too.

Want this modelled against your real deposit profile instead of estimated off a range? Get a cost model for your volume from LightningPay.

How does build vs buy actually compare on a first-year basis?

The build vs buy crypto payment gateway question usually gets framed as "vendor fees versus no fees." Wrong framing. Building doesn't remove the cost, it converts a variable cost into a fixed one. And fixed costs don't scale down when Q3 goes quiet.

What you're actually buying with headcount. A production-grade crypto deposit and withdrawal system needs node infrastructure or paid node providers on every chain you support. Hot and cold wallet architecture with real key management. Address derivation and deposit attribution.

Confirmation and reorg handling. A withdrawal approval workflow with proper controls. Blockchain analytics integration. Reconciliation and reporting your auditor will actually sign off. Monitoring and alerting.

Plus somebody awake at 3am when a chain forks, or a node silently drops twelve blocks behind and deposits stop crediting without a single error in the logs.

Realistically: two to three engineers to deliver over several months, then one to two allocated permanently for maintenance, new chain support and security work.

At European or North American loaded salary, that ongoing engineering commitment alone usually exceeds what a mid-volume operator would pay a vendor in fees. Before you've bought a single node subscription.

Where building genuinely wins. Very high volume, an existing crypto engineering team, and a business where payments infrastructure is strategic rather than supporting. Eight-figure monthly crypto volume with treasury already run in-house? The arithmetic can flip. It does happen. It just happens far less often than the pitch deck suggests.

Where it reliably loses. Any operator with game integrations, retention tooling or a platform migration on the roadmap. Every sprint spent on wallet infrastructure is a sprint not spent on something that earns. That opportunity cost never shows up in the build estimate. It never shows up in the post-mortem either, which is exactly why the same mistake keeps getting repeated.

Most operators end up on the same honest middle path: buy the gateway, own the treasury policy. Take a branded crypto payment gateway for iGaming operators for the infrastructure layer, and keep settlement currency, hedging and float decisions in-house where they belong.

What does a slow launch actually cost in lost deposits?

Here's the line finance leaves out. It's frequently bigger than every other line combined.

Work it backwards from your own numbers. Take current monthly deposit volume. Apply a conservative share that would either move to crypto or arrive through it.

For plenty of operators in crypto-forward or emerging markets that's 10-30% of deposits, and higher again for brands targeting regions where card acceptance is a coin flip. That's your monthly exposure.

An operator doing €2m in monthly deposits with a 15% crypto opportunity is leaving roughly €300,000 of monthly deposit volume on the floor. At a 4-6% net margin on deposits, call it €12,000-18,000 in GGR per month of delay.

A six-month in-house build that ships two months late has already burned more in foregone revenue than most white label integrations cost in a full year of fees.

There's a second-order cost, and it's the nastier one. Players who try to deposit in crypto and can't don't sit patiently. They deposit somewhere else. The acquisition spend you already committed converts for a competitor. You paid for the click. Someone else banked the player.

Which is why the realistic timeline to a live white label crypto gateway belongs in the cost conversation, not just the project plan. Time-to-first-deposit is a financial variable. Treat it like one.

How does instant Lightning settlement change the unit economics?

One capability is worth pricing on its own: instant Lightning settlement.

On-chain deposits sit in confirmation limbo before you can safely credit them. From your side that's funds you can't treat as settled, plus a working-capital gap you either fund yourself or pass to the player as a delay.

Lightning kills the window. The deposit is final in seconds. No float to carry, no confirmation-risk policy to draft, no "your deposit is pending, please wait" conversation in live chat.

It also flattens per-deposit network fee exposure. Lightning routing fees are a rounding error next to on-chain fees, and they don't spike when the mempool fills up.

For a high-frequency casino with a small average deposit, and I mean the €20-50 slots player topping up three times in one session, that's decisive.

When on-chain fees on a €25 deposit can swallow several percent, moving that flow to Lightning takes it from marginally viable to comfortably profitable. It also removes the scenario where a congestion spike quietly turns your smallest deposits loss-making, and nobody notices until the monthly report lands.

Final thoughts

Operators keep running the wrong comparison. 1.2% versus 1.8%. Licence fee versus percentage. That prices the gateway. It doesn't price the outcome.

What actually decides your crypto payments P&L is fully loaded cost per successful deposit: vendor fee, plus network fee, plus payout and treasury conversion, plus the support hours every broken deposit generates, plus the finance and audit uplift, plus the engineering time you spent here instead of on retention, plus the revenue you never earned while the integration sat half-finished.

All divided by a denominator that depends entirely on how well the cashier converts.

Two vendors quoting the same rate can land miles apart on that basis. The cheaper headline is frequently the more expensive gateway.

So stop asking for a rate card.

Ask vendors to model cost per successful deposit at your volume, your average deposit size and your coin mix, and to show their assumptions. If a vendor won't do that exercise with you, you've just learned something useful about how well they understand iGaming deposit flow.

To run the model against your own numbers, get a cost model for your volume from LightningPay.