How to Compare PSP Exchange Rates

How to Compare PSP Exchange Rates: A Fair Test for FX Spreads and Fees

At month-end, a treasury team tries to answer what looks like a simple procurement question:

Which payment provider gives us the better exchange rate?

One proposal contains a quoted FX markup.

Another settlement report contains an exchange rate but no separate FX charge.

A third provider describes its rates as competitive but does not specify a fixed spread.

The first quote was captured in the morning. The settlement transaction happened several hours later. One conversion went directly from the transaction currency into the merchant's settlement currency. Another passed through an intermediate balance currency before payout.

Putting those numbers in adjacent spreadsheet columns does not produce a comparison.

It produces three different economic events.

A defensible PSP FX comparison starts only after the merchant makes those events equivalent.

That means holding constant, or explicitly accounting for:

•currency pair;

•direction;

•source amount;

•conversion stage;

•timestamp;

•benchmark;

•explicit fees;

•settlement currency;

•payout currency;

•additional conversions;

•commercial volume tier.

Only then does it make sense to ask which provider costs less.

The key principle is simple:

Compare the same economic conversion scenario, not the rates providers happen to advertise.

The editorial brief makes this controlled-comparison methodology the article's core territory.

Why most PSP exchange-rate comparisons are invalid

Foreign-exchange comparisons fail surprisingly early.

The problem is usually not complicated mathematics. It is that the inputs describe different things.

The timestamps do not match

Foreign-exchange markets move.

A rate recorded at 10:00 and another recorded at 16:00 cannot be treated as though the market stood still between them.

Even public reference rates make timing explicit. The ECB generally publishes its euro foreign-exchange reference rates at around 16:00 CET, based on a daily central-bank concertation procedure that takes place earlier in the afternoon.

That makes a morning PSP quote versus an afternoon ECB reference rate a weak test of provider markup.

The quote directions differ

A provider may show:

USD per EUR

while another system records:

EUR per USD.

Both can describe the same market relationship.

They cannot be directly subtracted.

The amounts differ

A small test conversion and a large merchant settlement may receive different commercial treatment.

Enterprise FX pricing can also be contract-specific, so a public quote or small retail-style test should not automatically be treated as the price applicable to production merchant volume.

One figure includes fees and another does not

A provider can use:

•an explicit percentage conversion fee;

•a fixed charge;

•an embedded rate difference;

•some combination of these.

The nominal exchange rate is therefore only one input into the result.

The currency paths differ

A merchant can receive one quoted conversion while eventually paying for two.

For example:

EUR transaction proceeds

→ USD merchant balance

→ GBP payout

is not economically comparable with:

EUR proceeds

→ GBP settlement

unless both conversions in the first route are measured.

The brief therefore requires settlement and payout conversion to remain separate.

Start with the conversion path, not the headline rate

Before comparing any PSP exchange rates, write the currency path.

For example:

Customer transaction currency

→ processing currency, if different

→ settlement currency

→ merchant balance currency

→ payout currency

Every arrow that changes currency is a potential FX event.

Now identify each conversion separately.

For every arrow, record:

•Source currency

•Source amount

•Destination currency

•Destination amount

•Party performing or pricing the conversion

•Exchange rate

•Rate timestamp

•Benchmark

•Explicit fee

•Additional conversion-related charges

Without that map, two “exchange rates” may not even refer to the same stage of the payment flow.

Normalize the currency pair and direction

Suppose one system reports:

1 EUR = 1.10 USD

while another reports:

1 USD = 0.9091 EUR.

Those are approximately inverse representations of the same relationship.

A direct numerical comparison would be meaningless.

For procurement, standardize every observation as:

Destination currency per unit of source currency

If the merchant is converting EUR into USD:

EUR → USD

all provider and benchmark values should show:

USD received per EUR converted.

When systems cannot be normalized cleanly, skip the rate arithmetic and compare actual source and destination amounts instead.

That is often safer.

Define what the benchmark is supposed to tell you

A benchmark answers:

“What comparison point are we using?”

It does not necessarily answer:

“What price could we have executed at?”

Potential benchmarks include:

•central-bank reference rates;

•independent market-data benchmarks;

•contractual base rates;

•card/network conversion rates where relevant;

•contemporaneous executable quotes.

The ECB is especially explicit about this distinction. Its euro reference rates are published for information purposes, and the ECB strongly discourages their use as transaction rates. Its methodology notes that reference rates do not necessarily represent prices at which actual market transactions took place.

So instead of writing:

“The ECB rate was the true rate.”

write:

“The ECB rate was the benchmark used for this analysis.”

That is both more precise and more useful.

“Mid-market” needs the same discipline

The phrase mid-market rate is often used as though it identifies one universally executable price.

For merchant analysis, that is too loose.

The wholesale FX market has bid and ask prices. BIS defines the bid–ask spread as the difference between the price at which a customer sells a currency and the price at which a customer buys it.

A provider's deviation from a selected midpoint can be useful to measure.

But that deviation should not automatically be described as:

provider profit

or:

the wholesale bid–ask spread.

The merchant can defensibly say:

“Our actual conversion was X basis points away from benchmark Y at timestamp Z.”

That is a measurable statement.

Align the timestamps

This is the most common methodological failure in casual FX comparisons.

Suppose procurement records:

PSP quote: 10:02

and:

benchmark: 16:00.

The difference contains at least two possible effects:

•provider pricing;

•market movement between 10:02 and the benchmark observation.

Without separating them, the merchant cannot attribute the entire difference to the PSP.

The comparison should therefore use:

same timestamp

or:

the tightest practical observation window.

When perfect alignment is impossible, document the timing difference and reduce the confidence of the conclusion.

A simple confidence scheme

High confidence

Provider execution and benchmark are essentially contemporaneous.

Medium confidence

Small timing gap in relatively normal market conditions.

Low confidence

Material timing gap or uncertain execution timestamp.

Do not create basis-point precision that the source data cannot support.

Compare destination amounts, not just rates

Once pair, direction and timestamp are normalized, translate the comparison into money.

Suppose, purely illustratively:

Source amount: €250,000

Benchmark: 1 EUR = 1.1200 USD

Provider conversion rate: 1 EUR = 1.1144 USD

Benchmark destination amount

€250,000 × 1.1200

=

$280,000

Provider destination amount

€250,000 × 1.1144

=

$278,600

Difference

$280,000 − $278,600

=

$1,400

Effective difference to benchmark

$1,400 ÷ $280,000

=

0.50%

or:

50 basis points

The advantage of amount-level analysis is that finance immediately sees what the rate difference means economically.

Add every explicit FX charge

Now suppose the provider also charges an explicit conversion fee.

The comparison is no longer:

benchmark rate versus provider rate.

It is:

Benchmark destination value

versus:

Final merchant destination value after conversion-related charges

A useful formula is:

All-in effective conversion rate

Final destination amount after applicable FX charges

÷ source amount

Then compare that all-in result against the chosen benchmark.

The brief specifically identifies final destination amount as more useful than nominal headline spread when comparing providers.

“No FX fee” is not the same as “no FX cost”

This is one of the strongest points in the original draft and should remain.

A provider can genuinely charge:

0% separately itemized FX fee

while applying a conversion rate away from the selected benchmark.

Another provider can charge:

an explicit FX fee

while giving a conversion rate closer to that benchmark.

Either could produce the better all-in outcome.

Therefore:

No explicit FX fee ≠ zero economic FX cost.

But the inverse caution matters too:

Deviation from a public benchmark ≠ proof of an undisclosed fee.

A public benchmark is not necessarily executable and does not reveal the provider's entire economics.

Use:

observed benchmark difference

or:

effective FX cost

unless contractual evidence establishes a particular markup.

Do not stop at settlement if payout converts again

A merchant may evaluate an apparently attractive settlement rate and miss what happens next.

Suppose payment proceeds become:

EUR → USD at settlement

then:

USD → GBP at payout.

If the merchant economically needs GBP, the all-in provider comparison needs both conversions.

For the first conversion:

•EUR amount;

•USD amount;

•rate;

•timestamp;

•benchmark;

•fee.

For the second:

•USD amount;

•GBP amount;

•rate;

•timestamp;

•benchmark;

•fee.

Only then should procurement compare that path against a provider capable of a direct:

EUR → GBP

route.

This does not mean direct conversion is always available or cheaper.

Operational, banking, provider or regulatory constraints can make an intermediate currency necessary. The brief specifically warns against assuming every multi-step path is avoidable.

Settlement currency can change which provider is cheaper

Consider two commercial arrangements.

One offers an attractive EUR→USD conversion but automatically converts all EUR proceeds to USD.

Another permits the merchant to retain EUR.

For a merchant whose operating expenses are substantially in EUR, the second arrangement may avoid a future:

USD → EUR

conversion.

The narrow first conversion could look worse while the total currency path is better.

That is why PSP procurement must record:

•settlement currency;

•balance currency;

•payout currency;

•automatic conversion rules.

The article's purpose is not to argue that merchants should always hold multiple currencies. Additional balances can create reconciliation, treasury and operational complexity.

The point is that settlement architecture is part of FX pricing comparison.

Public pricing is not the same as negotiated merchant pricing

A public rate page can be useful for discovery.

It should not automatically become the procurement model.

Merchant FX economics may depend on:

•currency pair;

•volume;

•merchant size;

•contract;

•country;

•conversion timing;

•settlement structure.

The brief explicitly requires public pricing, negotiated merchant terms and illustrative examples to remain distinct.

For an established merchant, the stronger evidence hierarchy is usually:

actual merchant contract

→ actual conversion records

→ settlement reports

→ provider proposal

→ public pricing

rather than the reverse.

One spot test is not enough

Suppose one provider produces the better GBP→USD conversion today.

That does not prove it will be cheaper across the merchant's actual portfolio.

FX comparisons should be repeated across:

Time

Test multiple observations rather than one market moment.

Currency pairs

Test the pairs the merchant actually converts.

Amounts

Use sizes representative of production flows.

Commercial tiers

Use the merchant's actual or proposed contract tier.

Conversion stages

Settlement and payout may have different pricing.

The brief explicitly calls for multiple timestamps, multiple currency pairs and realistic volumes to avoid overinterpreting a single spot observation.

Pair-level results matter more than a global headline

A provider may be attractive for:

EUR → USD

and materially less attractive for:

PHP → USD

or:

THB → EUR.

The reason could involve provider commercial pricing, market conditions, settlement availability or other currency-specific factors.

Wholesale liquidity can influence FX economics, but a merchant should not infer its PSP spread directly from wholesale market liquidity.

So instead of reporting:

“Provider X costs 42 bps for FX.”

report:

EUR→USD: ___

GBP→USD: ___

JPY→USD: ___

and weight them by merchant conversion volume.

Weight by actual economic volume

A simple average can make the wrong provider look cheap.

Suppose a merchant observes hypothetical all-in benchmark differences of:

•EUR→USD: 25 bps

•GBP→USD: 35 bps

•PHP→USD: 120 bps

Taking the arithmetic mean gives:

60 bps.

But if 85% of conversion value is EUR→USD and only 1% is PHP→USD, that average is economically misleading.

Instead calculate:

Weighted effective FX cost

Total measured FX cost across included conversions

÷ total benchmark-converted value

The cluster brief explicitly requires weighting by economic volume rather than averaging percentages equally across pairs.

Translate basis points into annual expense

Procurement needs dollars, euros or another reporting currency—not just basis points.

For each pair:

Expected annual FX expense

Expected annual conversion volume

× measured all-in FX cost rate

Suppose:

Eligible annual conversion volume = $200 million equivalent

and a proposed commercial change improves all-in cost by an illustrative:

15 bps = 0.15%

Gross annual difference:

$200,000,000 × 0.0015 = $300,000

That is a gross model result, not guaranteed savings.

Any incremental:

•account cost;

•integration;

•settlement;

•treasury;

•operational overhead

should be deducted before calling it net benefit.

Rate consistency matters as well as average price

Average FX cost does not tell the whole story.

Treasury may also care about:

variance.

A provider with a slightly higher average cost but consistent execution may be easier to forecast than one whose observed results vary significantly.

Track:

•median benchmark difference;

•volume-weighted average;

•variance or range;

•high-percentile observations;

•number of conversions;

•tested volume.

This converts “the rate looked good when we checked” into a repeatable procurement metric.

What “real-time rates” tells you—and what it does not

“Real-time” describes a timing characteristic.

It does not by itself establish:

•benchmark methodology;

•spread;

•markup;

•absence of an explicit charge;

•competitiveness.

A real-time provider rate can still contain commercial pricing.

Likewise, a periodically fixed rate can be commercially attractive.

The procurement question is:

What final economic result does the rate mechanism produce for our actual conversions?

What “competitive rates” tells you

Almost nothing quantitative.

The correct follow-up is not necessarily:

“What percentage do you charge?”

It is:

“Show us the rate methodology and the resulting all-in destination amount for these controlled scenarios.”

Ask for:

•base/reference-rate methodology;

•markup or spread methodology if contractually defined;

•exact timing;

•explicit fees;

•currency-pair differences;

•volume tiers;

•settlement currencies;

•payout currencies;

•automatic conversions.

What “transparent pricing” should mean in procurement

Transparency should be testable.

A merchant should be able to reconstruct:

source amount

→ rate

→ explicit conversion charges

→ destination amount.

And, ideally, identify:

when the rate was determined

and:

which conversion stage it belongs to.

This does not require every provider to expose its internal wholesale economics.

It requires enough information for the merchant to verify its own economic result.

Reporting quality is part of the commercial comparison

Imagine two PSPs produce nearly identical all-in FX cost.

One gives finance:

•source amount;

•source currency;

•destination currency;

•applied rate;

•conversion time;

•fee;

•settlement identifier.

The other provides only a final payout amount.

The direct FX price might be similar.

The operational value is not.

Clearer data can reduce:

•reconciliation work;

•investigation time;

•procurement uncertainty;

•model error.

That should be captured separately in the scorecard rather than pretending it is another basis-point fee.

The PSP FX Rate Comparison Scorecard

For every provider and tested conversion, record:

Field

What to capture

Currency pair

Source → destination

Source amount

Production-relevant size

Direction

Explicitly normalized

Conversion stage

Settlement / balance / payout

Provider quote/execution timestamp

Exact where available

Benchmark source

Named reference

Benchmark timestamp

Aligned to execution

Provider rate

Normalized direction

Explicit FX fee

Amount/rate and fee basis

Gross destination amount

Before separate fee where relevant

Final destination amount

After included FX charges

Benchmark destination amount

Controlled comparison

Effective benchmark difference

% / bps

Settlement currency

Actual

Payout currency

Actual

Additional conversion

Yes / no

Volume tier

Actual/proposed

Rate mechanism

Locked / settlement-time / other verified method

Reporting transparency

High / medium / low

Observation confidence

High / medium / low

This is the practical asset that distinguishes this article from the separate What Is an FX Spread? article. The latter explains the arithmetic; this one establishes a repeatable provider procurement procedure.

Use an all-in effective rate

For one conversion:

All-in effective conversion rate

Final destination amount after included FX charges

÷ source amount

Then measure it against the normalized benchmark.

If:

benchmark result = 150,000 destination-currency units

and:

actual all-in result = 149,250

then:

difference = 750

and:

Effective cost to benchmark

750 ÷ 150,000 = 0.50% = 50 bps

Again, the benchmark must be named.

Do not simply label the result:

“provider's 50-bp profit margin.”

A stronger procurement test uses repeated observations

A useful PSP comparison program can operate in three rounds.

Round 1 — controlled proposal test

Provide each PSP the same:

•currency pair;

•direction;

•source amount;

•expected monthly volume;

•settlement currency;

•payout currency;

•merchant profile.

Ask them to price the scenario.

Round 2 — parallel observation

Where commercially and operationally possible, collect comparable live or pilot data over multiple timestamps.

Record actual conversion outcomes.

Round 3 — annualized model

Apply each provider's observed or contractually agreed economics to the merchant's actual currency mix.

This produces:

expected annual FX cost by provider

rather than:

one attractive screenshot.

Do not ignore refunds, but keep them out of the base test

Refund FX belongs to a separate lifecycle event.

A provider may:

•reuse an original conversion treatment;

•apply a later rate;

•alter merchant settlement differently.

That can matter to the total provider economics, especially for merchants with high refund volumes.

But do not combine refund FX into the basic exchange-rate comparison without identifying it separately.

The dedicated refund article should own the detailed calculation.

Likewise, do not confuse conversion cost with FX exposure

A PSP exchange-rate comparison asks:

What did the conversion cost?

Treasury exposure asks:

How did the economic value of an unconverted foreign-currency position change over time?

Those are related but different.

A provider that lets the merchant hold EUR may reduce immediate conversion but create an open EUR position relative to the merchant's reporting currency.

That is not automatically a loss.

It is exposure.

The Managing Payment FX Exposure article should own that decision.

The cheapest observed conversion is not automatically the best PSP

Consider what happens if the lowest measured spread comes with:

•mandatory conversion into one settlement currency;

•limited payout currencies;

•another downstream conversion;

•poorer reporting;

•more reconciliation;

•reduced operational flexibility.

The merchant's decision should be based on total payment and treasury economics.

The broader editorial standard for this cluster is explicitly not to obtain the public benchmark at all costs. The objective is to minimize unnecessary currency-conversion cost while maintaining settlement, liquidity, reporting and operational simplicity.

Three procurement scenarios

Scenario 1: The advertised spread winner loses after fees

The proposal with the tightest quoted spread also charges a separate conversion fee.

After both are applied, another provider produces more destination currency.

Decision rule: rank the final all-in destination amount, not the quoted spread.

Scenario 2: The settlement winner loses after payout

One PSP gives a better EUR→USD settlement rate.

But treasury ultimately needs GBP and the balance must later convert USD→GBP.

Another setup settles directly into GBP.

Decision rule: compare the complete currency path rather than one favorable conversion stage.

Scenario 3: The cheapest spot test is not the cheapest portfolio

One PSP wins a single EUR→USD observation.

Across three months of production-like testing, another performs better on the merchant's largest currency pairs and has lower volume-weighted all-in cost.

Decision rule: use repeated, volume-weighted observations rather than declaring a winner from one timestamp.

PSP exchange-rate comparison checklist

Before signing off on an FX comparison, confirm that:

•The source currency is identical.

•The destination currency is identical.

•Rate direction is normalized.

•Source amounts are comparable.

•Provider timestamps are known.

•Benchmark timestamps are aligned.

•The benchmark source is named.

•The benchmark is not being misrepresented as an executable rate.

•Explicit FX charges are included.

•Final destination amount is calculated.

•Settlement currency is identical or its difference is modeled.

•Payout currency is identical or its difference is modeled.

•Any second conversion is included.

•Actual merchant pricing tier is used.

•Multiple observations are tested.

•Major currency pairs are weighted by real economic volume.

•Reporting quality is evaluated separately.

•Gross rate improvement is not automatically described as net savings.

If those conditions are not satisfied, the merchant may still have useful FX data.

It does not yet have a fair PSP comparison.

Legal, financial and treasury note

This article provides general information and does not constitute legal, tax, accounting, investment, treasury or financial advice. Merchants should verify current currency rules, provider terms, settlement restrictions and accounting treatment with qualified advisers, banks, payment providers and relevant authorities.

I’ve rebuilt the piece around the new Editorial Brief 7: How to Compare PSP Exchange Rates, while preserving the strongest parts of your draft. The biggest editorial change is methodological: the article now treats PSP FX procurement as a controlled comparison of complete conversion outcomes, rather than a contest between advertised spreads. It also follows your new instruction to avoid another “Provider A vs Provider B” opening and closes with procurement scenarios plus a checklist instead of a generic “Final Recommendation.”

I also tightened one substantive point from the reference draft: a public “mid-market” or central-bank rate is useful as a benchmark, but should not be described as the rate a merchant necessarily “could be getting.” The ECB explicitly says its euro reference rates are for information purposes and strongly discourages their use as transaction rates; those rates are averages and do not necessarily reflect executable market transactions.

The most important edit from your original foundation is the replacement of “the mid-market/interbank rate is the actual rate currencies trade at and what merchants could be getting” with a more defensible benchmark formulation. Public reference rates are extremely useful for measurement, but official sources themselves caution that they are not necessarily executable transaction rates.

I would also hold the Moment Treasury paragraph from publication until its product name, capabilities, settlement-currency coverage and “190+ markets” claim can be confirmed internally. The rest of the supplied foundation translates very well into the controlled-comparison framework required by the new brief.

For this article, a safe transition would be:

For merchants working across multiple currencies, provider evaluation should include not only the rate applied at conversion but also the settlement and payout structure surrounding it. A payments provider that gives finance enough data to trace source currency, settlement currency, payout currency and conversion outcomes can make all-in FX economics easier to measure and compare.

The submitted foundation includes:

“Moment Treasury is built around giving merchants visibility into multi-currency settlement”

and:

“merchants operating across 190+ markets.”

Do not publish those claims until internal documentation confirms:

•current product name;

•product availability;

•supported settlement currencies;

•supported payout currencies;

•multi-currency balances;

•FX methodology;

•reporting functionality;

•relevant geographic coverage;

•approved “190+ markets” wording;

•whether visibility into conversion rates/spreads is a supported product capability.

SEO package

H1: How to Compare PSP Exchange Rates: A Fair Test for FX Spreads and Fees

Meta description: Learn how to compare payment-provider exchange rates using the same currency pair, amount, timestamp, benchmark, settlement path and all-in destination amount.

URL slug: /compare-psp-exchange-rates

Secondary keywords: compare PSP FX rates; payment provider exchange rates; PSP FX comparison; merchant foreign exchange rates; compare payment FX fees; PSP exchange rate markup; payment provider FX spread; FX pricing comparison.

Related entities: foreign exchange, FX spread, exchange-rate benchmark, settlement currency, payout currency, effective exchange rate, basis points, treasury, multi-currency settlement.

Article excerpt: Comparing PSP exchange rates fairly requires more than lining up quoted spreads. Merchants need to normalize currency pair, direction, amount, timestamp, fees and settlement path—then compare the final destination value under the same economic scenario.

Structured data: Article.

Supporting visual

PSP FX Rate Comparison Scorecard

Visual concept:

NORMALIZE

Currency pair

→ Direction

→ Amount

→ Timestamp

↓

PRICE

Benchmark

→ Provider rate

→ Explicit FX fee

↓

FOLLOW THE MONEY

Settlement currency

→ Balance currency

→ Payout currency

→ Additional conversion?

↓

COMPARE

Final destination amount

→ Effective benchmark difference

→ Annual weighted FX cost

Primary callout

A rate comparison is valid only when the economic conversion scenario is the same.

Alt text

PSP FX Rate Comparison Scorecard showing how merchants normalize currency pair, direction, amount, timestamp, benchmark, fees, settlement currency and payout currency before comparing final FX cost.

Measurement plan

Track for each provider:

All-in effective FX cost

Benchmark difference in bps

Absolute FX cost

Variance over repeated observations

Cost by currency pair

Volume-weighted effective FX cost

Expected annual FX expense

Additional conversion count

Reporting-confidence score

For procurement, the primary commercial output should be:

Expected annual all-in FX expense

based on the merchant's actual currency-pair mix and realistic conversion volume.

Authoritative research foundation

European Central Bank: Useful as a clearly defined reference-rate source and, critically, as evidence that a public reference rate is not automatically an executable merchant transaction rate.

Bank for International Settlements: Useful for keeping the conventional market meaning of bid–ask spread separate from a merchant's provider-to-benchmark cost calculation.

FX Global Code / Global Foreign Exchange Committee: The current Code was updated in December 2024 and promotes robust, fair, liquid, open and appropriately transparent wholesale FX markets. It can support the article's wider emphasis on repeatable, transparent comparison practice without suggesting that it dictates PSP merchant pricing.

Cluster sameness audit

Field

This article

Opening

Treasury/procurement spreadsheet containing incomparable rate evidence

Primary problem

False provider comparison caused by non-normalized inputs

Source/destination currency

Variable; provider test must normalize them

Conversion stage

Controlled and explicitly identified

Converting party

PSP/provider/bank actually pricing the conversion

Benchmark

Defined comparison source with timestamp

Format

Controlled PSP FX benchmark

Evidence anchor

Rates from different times or paths cannot be fairly compared

Editorial position

Compare final destination amounts under controlled conditions

Merchant decision

Which provider creates the lowest sustainable all-in cost for the merchant's actual FX portfolio

Original asset

PSP FX Rate Comparison Scorecard

Primary metric

Volume-weighted all-in effective FX cost

Ending

Three procurement scenarios + checklist

CTA

Build repeatable controlled observations rather than compare marketing pages

This keeps the article distinct from What Is an FX Spread?, which owns spread mechanics; Hidden FX Markups, which owns forensic detection; and FX Costs by Market, which owns currency/market-specific economics.

v1

How to Compare PSP Exchange Rates

Two PSPs can both advertise "no FX fees" and still cost a merchant meaningfully different amounts on the same cross-border transaction. That's because the fee isn't always the part that costs money — the exchange rate itself often is, and it's the part merchants are least equipped to compare, because most PSPs don't make it easy to see what rate you're actually getting versus what you could be getting.

This is a genuinely difficult comparison to make well, but it's also one of the more consequential ones for any merchant processing meaningful cross-border volume — the difference compounds on every transaction, not just the occasional one.

Why This Comparison Is Harder Than It Looks

"No FX fee" doesn't mean no FX cost. A provider can genuinely charge zero separate fee for currency conversion and still make money on FX by applying an exchange rate that's worse than the mid-market rate. The margin is baked into the rate itself rather than itemized as a line item, which makes it invisible unless you specifically go looking for it.

Quotes are rarely apples-to-apples. One PSP's pricing page might show a rate as of a specific moment; another might quote a typical spread as a range; a third might not disclose it at all until you're processing live volume. Comparing these directly is close to meaningless without normalizing them to the same reference point.

Rates move constantly. Currency markets fluctuate throughout the day, so a rate comparison done at 9am isn't automatically valid at 3pm. Any comparison needs to account for this rather than treating a single quoted rate as a fixed, permanent number.

Blended pricing can hide FX costs inside a bundled rate. Some PSPs quote a single all-in processing rate that includes FX conversion baked in, rather than breaking out acquiring cost, scheme fees, and FX margin separately. That's not necessarily worse, but it makes direct comparison to a provider using interchange-plus-style itemized pricing much harder without asking for a breakdown.

What Actually Makes Up the Rate You're Paying

Every FX-inclusive transaction has, in principle, two components:

•The mid-market (interbank) rate — the actual rate at which currencies are trading between banks at that moment, with no markup. This is the reference point everything should be measured against.

•The spread or markup — the difference between the mid-market rate and the rate you're actually given. This is where the PSP's FX margin lives, whether or not it's disclosed as a separate fee.

The spread is the number that matters for comparison purposes. A provider charging a visible 1% FX fee on top of the true mid-market rate can be cheaper than a provider charging "0% FX fee" but quietly applying a rate 2% off mid-market.

How to Actually Compare Providers

Get the rate and the mid-market reference at the same timestamp. Ask each provider (or check their live transaction data if you're already processing with them) what rate was applied to a specific transaction, then compare it to a public mid-market rate source at that exact time. The percentage difference between the two is the real number to compare — not the marketing claim about fees.

Ask directly what the FX margin is, in basis points, over mid-market. A provider that can answer this clearly and consistently is giving you something you can actually compare. A provider that can only point to a marketing page saying "competitive rates" or "no hidden fees" is not.

Look at your actual settlement reports, not the advertised rate. If you're already live with a provider, the most reliable comparison isn't the quote — it's the historical difference between what you were charged in presentment currency and what actually landed in your settlement currency. This tells you the effective rate you received, which is the only number that actually matters.

Test with real transaction volume across your top markets, not a single sample. FX spreads can vary by currency pair — a provider might be competitive on EUR/USD and considerably worse on a less commonly traded pair like THB or PHP. A comparison based on one currency pair doesn't tell you much about your overall cost across a diversified market footprint.

Check whether presentment currency and settlement currency create a double conversion. If a transaction is presented in one currency, converted to a different intermediate currency, and then converted again to your settlement currency, you can absorb two spreads instead of one. Ask specifically whether your transactions pass through an intermediate currency, and whether local acquiring in a given market avoids that extra step.

Ask about rate consistency, not just the rate itself. Some providers offer a locked or guaranteed rate for a window of time; others apply the live market rate at the moment of settlement, which introduces variability a merchant has to account for separately in margin planning. Neither is inherently better, but you need to know which one you're getting to compare costs meaningfully.

What to Watch For in Marketing Language

•"No FX fees" — ask what rate is actually being applied relative to mid-market. This phrase says nothing about the spread.

•"Competitive rates" — not a number. Ask for the actual margin in basis points.

•"Transparent pricing" — ask to see it broken down by component: acquiring, scheme fees, and FX margin listed separately, not bundled into one number that can't be decomposed.

•"Real-time rates" — this describes timing, not cost. A real-time rate can still carry a real-time markup.

A Practical Approach

Rather than trying to negotiate FX pricing in the abstract, request a like-for-like comparison: the same currency pair, the same transaction size, the same day, compared against a neutral mid-market benchmark. Do this for your two or three highest-volume corridors specifically, since that's where the spread differential will have the largest actual dollar impact — a slightly worse rate on a rarely used currency pair matters far less than the same gap on your largest market.

Moment Treasury is built around giving merchants visibility into multi-currency settlement rather than treating FX as a black box baked into a blended rate. For merchants operating across 190+ markets, the goal is making the actual cost of currency conversion something that can be checked and compared, rather than something that only shows up as a gap between the price a customer saw and the amount that eventually settles.