Every fee comparison stops at the trading schedule. But before a single order is placed, money has to enter the system, and afterwards it has to leave, and both of those movements have a cost that appears on no schedule and gets quietly deducted from returns.
Trading fees are published, comparable and competed on. Ramp costs are none of those things. They vary by payment method, by currency, by corridor, by amount, sometimes by the day, and they are frequently expressed in a way that makes them hard to compare with anything else. Which is precisely why they persist at levels trading fees left behind years ago.
The asymmetry is worth stating plainly. A trader will compare fee schedules to the decimal place and then move money in through a method costing several times the entire round trip they were optimising. The attention goes where the numbers are published, and the ramp is where the numbers are not.
This article is about measuring the part nobody publishes clearly. Our own trading schedule is below and it is the same for everybody, because that part we do publish. The ramp costs described here are general to the industry and depend on your method, your currency and your bank far more than on any single platform.
| Market | Maker | Taker | What it means |
|---|---|---|---|
| Buying and selling assets outright | 0.100 % | 0.100 % | 0.0100 % |
| Leveraged positions on contracts | 0.020 % | 0.060 % | 0.0060 % |
The first is an explicit deposit or withdrawal fee, a stated amount or percentage. This is the honest version and the easiest to compare, because it is a number you can read before committing. It is also, on most established platforms, the smallest of the four.
The second is the conversion spread. If you fund in one currency and the platform quotes in another, a rate is applied, and the difference between that rate and the interbank rate is a cost. It is never presented as a fee, it is presented as a rate, and the two are very different in how visible they are.
The third is the payment network's own charge, which reaches you through your bank or card issuer rather than through the platform. Cross-border transfer fees, correspondent bank deductions and card cash-advance treatment all sit here, and none of them appear anywhere in the platform's interface. The fourth is time, which is a cost whenever the market moves while your funds are in transit.
Four costs: the stated fee, the conversion spread, your bank's own charge, and the days your money spends in transit. Only the first appears in the interface.
Measuring a conversion cost requires a reference, and the reference is the mid-market rate: the midpoint between what buyers and sellers are quoting for that currency pair at that moment. It is published freely and it is what the currency actually costs when nobody is taking a margin.
Take the rate you were given, take the mid-market rate at the same moment, and the percentage difference is your conversion cost. Do it once for each method you use and the results are frequently startling, because a spread that looks like a rounding difference on the quoted number is a substantial percentage on the amount converted.
The practical implication is that funding in the currency the platform quotes in, where that is possible, removes an entire cost category. It is not always possible and it is not always convenient, but it is worth knowing which of your movements involve a conversion and which do not, because the answer is not always what you would assume.
Bank transfers are generally the cheapest and the slowest, and within them the cost depends heavily on whether the transfer stays inside a payment area or crosses one. A domestic or single-area transfer is frequently free or nearly so; the same transfer crossing borders can pass through correspondent banks that each take a deduction you never see itemised.
Cards are the fastest and among the most expensive, and there is a specific trap. Some issuers treat a purchase of digital assets as a cash advance rather than a purchase, which triggers a separate fee and starts interest accruing immediately with no grace period. The platform's fee is not the whole cost and the rest arrives on a statement later.
Payment applications and third-party processors sit in between and vary enormously. The one consistent pattern is that convenience is priced: methods that settle in seconds and require no preparation cost more than methods that take days and require a bank relationship. That trade-off is legitimate and it is worth making consciously rather than by default.
Many traders never move fiat at all after their first entry. They hold value in a dollar-referenced stablecoin between positions, move it between venues on a chain, and only convert back to bank money when withdrawing for real-world use. The costs of that route are different, not absent.
Moving on a chain costs a network fee, which varies by chain and by congestion and which is unrelated to the amount being moved. That last part matters: the same transfer costs the same whether it carries a small amount or a large one, which makes chain movement expensive for small sums and cheap for large ones. The cost structure is the opposite of a percentage fee.
There is also the question of what the stablecoin itself is, and that is a genuine risk rather than a cost. A dollar-referenced token is a claim on an issuer or a mechanism, and the reasons a claim can fail to be worth a dollar are covered separately. Treating that route as free of risk because it is free of fiat is a mistake.
The first movement into the system is systematically the most expensive one, for reasons that are structural rather than predatory. It typically involves a conversion, a payment method chosen for speed rather than cost, and an amount small enough that fixed costs weigh heavily. Everything after it can be optimised; the first one usually is not.
This has a consequence people miss. Somebody who enters with a small amount, trades it, and withdraws, may find the ramp costs on both ends exceeded everything the trading produced, and conclude that trading does not work. The trading may have been fine and the round trip through the banking system was the problem.
The structural answer is to make ramp movements fewer and larger rather than frequent and small, when the costs are fixed rather than proportional. That is a different discipline from optimising trade size, and it operates on a completely different timescale.
Almost everybody thinks about how to get money in and almost nobody thinks about how to get it out until they need to. That asymmetry produces bad outcomes, because withdrawal is where the friction is: verification requirements, limits, processing times and occasionally a rejected transfer that has to be redone.
The practical test is worth doing early. Move a small amount out, all the way to your bank account, before you have a reason to need it urgently. You will learn what your limits are, how long it takes, what it costs, and whether anything about your account needs attention. Learning that during a market event is learning it at the worst possible moment.
There is also a currency question on the way out. Withdrawing in a currency the platform holds and converting at your bank produces a different total cost from converting on the platform and withdrawing in your own currency, and which is cheaper depends entirely on the two spreads involved. It is worth measuring once rather than assuming.
Ramp costs come in two shapes and they imply opposite behaviour. A fixed fee, whether stated as an amount or arising from a network charge, weighs on small movements and disappears on large ones. A proportional fee weighs equally on all sizes and cannot be diluted by moving more at once.
Knowing which shape you face determines the correct strategy entirely. Facing fixed costs, movements should be as large and as infrequent as your needs allow. Facing proportional costs, size makes no difference and the only lever is choosing a cheaper method or a cheaper corridor.
Most real situations are a mixture: a fixed network fee plus a proportional conversion spread plus a fixed bank charge. Working out the mixture for your own path is an afternoon of arithmetic that pays for itself the first year, and it is arithmetic nobody will do for you because the numbers come from three different institutions.
The method is simple and almost nobody does it. Take one movement, record the amount that left your bank account, and record the amount that appeared as tradeable balance. The difference, as a percentage of the first number, is your true entry cost, and it includes every deduction from every party in the chain whether or not any of them itemised it.
Do the same on the way out: the balance that left the platform against the amount that arrived in your account. Add the two percentages and you have the cost of a complete round trip through the banking system, which is the number to compare against what your trading actually produces.
Traders who do this once tend to change something immediately, because the number is usually larger than they expected and it is entirely under their control. Unlike market movement, ramp cost is a fixed property of the path you chose, and choosing a different path changes it permanently.
The amount that left your account against the amount that became tradeable. That difference is the whole entry cost, itemised or not.
In order of typical impact: eliminate any conversion you do not need, by funding in the currency the platform quotes in where that is available to you. Then check whether your card issuer treats digital asset purchases as cash advances, because if it does, that single fact may dominate everything else on the list.
Then consolidate movements if your costs are fixed rather than proportional, and test a withdrawal end to end before you need one. Four checks, each of them a one-off, and together they typically account for most of what a regular trader loses to ramps over a year.
None of this involves changing how you trade, which is why it is worth doing first. It is the part of the cost structure that is fully knowable in advance and fully within your control, which is not true of anything that happens after the order is placed.
Because they depend on your payment method, currency, corridor and bank rather than on the platform alone, so a single published number could not be accurate. The consequence is that they are hard to compare, which is why they persist at levels trading fees left behind years ago.
Four places: an explicit deposit or withdrawal fee, the conversion spread if your currency differs from the quote currency, your own bank's or card issuer's charge which never appears in the platform interface, and the days your money spends in transit while the market moves.
Compare the rate you were given against the mid-market rate at the same moment, which is published freely. The percentage difference is your cost. A spread that looks like a rounding difference on the quoted rate is frequently a substantial percentage of the amount converted.
Bank transfers are generally cheapest and slowest, and cheaper still when they stay inside a single payment area rather than crossing borders. Cards are fastest and most expensive, with a specific trap: some issuers treat digital asset purchases as cash advances, adding a fee and immediate interest.
It has a different cost structure rather than no cost. A chain transfer costs a network fee unrelated to the amount, so it is expensive for small sums and cheap for large ones, the opposite of a percentage fee. The stablecoin itself also carries issuer risk, which is a risk and not a fee.
The first movement typically involves a conversion, a method chosen for speed rather than cost, and an amount small enough that fixed costs weigh heavily. Somebody entering small, trading, and withdrawing can find the ramp costs on both ends exceeded what the trading produced.
Yes, and almost nobody does. Move a small amount all the way to your bank account before you urgently need to, so you learn your limits, the processing time, the cost and whether your account needs attention. Learning that during a market event is the worst possible moment.
Record what left your bank account and what appeared as tradeable balance; the percentage difference is your entry cost including every deduction from every party. Do the same outbound, add the two, and compare that round trip against what your trading actually produces.