We use cookies to understand how the site is used (analytics and session replay). See our privacy policy.

Home/Guides/FX cost benchmarking
Guides · The standard

What is independent FX cost benchmarking?

Updated July 20268 min readReviewed by Just's benchmarking team
The short answer

Independent FX cost benchmarking measures what your bank charges you when you buy or sell foreign currency. It scores each trade against two references: the real market rate the bank could actually have traded at (never the interbank mid, a midpoint price nobody can trade on), and a set of similar companies. It puts a number, in bps and in money, on a cost that never appears as a line item, and it gives you the evidence to negotiate it down. Independent means paid by you, never the bank.

What it is

Every time a company converts currency, the bank adds a markup. That markup is the margin. It is folded into the single all-in rate on your trade, so it never appears as a fee. Benchmarking measures that margin, trade by trade, and asks one question: is it fair? The answer only means something against a reference, and independent benchmarking uses two: the rate the bank could actually have traded at, and what comparable companies pay for the same currencies and how far ahead each trade settles (the tenor).

Independent FX cost benchmarking: measuring a company's bank FX pricing against the real market rate at the moment of the trade, and against a group of similar companies, to establish whether the pricing is fair. Paid by the company, never the bank.

Why the cost is invisible today

The margin sits inside the all-in rate on your trade confirmation, so no invoice, statement or system will ever show it. On top of that, the information is one-sided: your bank sees what every one of its clients pays, and you see only your own rate. Told that your pricing is competitive, you have no independent number to check it against.

The make-up of that one all-in number makes the problem concrete.

Why independent matters

The question is whether the bank's pricing is fair, so the answer cannot come from the bank, and it cannot credibly come from anyone the bank pays. A benchmark can hold a price to account only when the party running it sits on the customer's side of the table. Just is paid by you, never the bank, so our only incentive is your result. We take no money from banks, no share of your trading, and we never route your trades.

The mid versus the real market rate

Most cost analysis compares your rate to the interbank mid: the midpoint between the price a bank will buy at and the price it will sell at. The gap between those two prices is the spread, and nobody can actually trade at the midpoint, so measuring against it produces a number your bank can wave away. Independent benchmarking scores every trade against the real market rate: a rate you could actually have traded at, for your trade size and direction (whether you were buying or selling), at the moment you traded.

Just's reference rate is a proprietary blend built from independent market-data feeds, live bank quotes, and several electronic trading venues (ECNs). The sources are kept confidential by contract, and the reference rate used to score each trade is shown on every scored trade, so every number can be traced back to what it was measured against. Most trades settle right away at today's rate, a spot trade; a forward settles on a future date instead. For forwards, the future-date adjustment is calculated across both standard and odd dates out to 3-4 years, so a 47-day forward is scored as precisely as a standard 1-month one.

Real market rate: a rate you could actually have traded at, for your trade size and direction, at the moment you traded. The margin is the gap between what you paid and that rate.

Similar companies

A trade scored against the real market rate gives you a margin in bps. A margin on its own is still hard to judge, so the second reference is a group of similar companies: what companies with comparable trading pay for the same currency pair, size, and settlement date. Just's benchmark is built on hundreds of companies, millions of benchmark trades and $2tn+ of total trade value measured. Customers have saved over $75M to date. As of mid-2026, measured across the Just benchmark network (counting basis).

Where there are too few similar companies to support a conclusion, no comparison is shown. Against that group, your number becomes a position: fair, or overpaying, and by how many bps, which converts directly into money on your trading volumes.

How it works in practice

01MeasureEvery trade scored against the real market rate at the moment of the trade.
02BenchmarkYour margin against similar companies: same currency pair, size, settlement date.
03NegotiateThe evidence becomes a bank-ready case. The conversation stays yours.
04MonitorNew trades scored as they land, so margins creeping back up are caught in weeks.

Scored, a single trade looks like this.

SCORED TRADE · EUR/NOK SPOTIllustrative
Trade valueEUR 5,000,000
Rate you were charged11.5579
Real market rate at that moment11.5420
Margin13.8 bps
Cost of this tradeNOK 79,500 · ≈ EUR 6,900

One trade is a data point. A year of trades, scored the same way, is a cost line you can govern.

See the full methodologyHow the reference rate is built, how deep the peer group goes, and the audit trail behind every scored trade.
Read the method

What changes when you measure

Savings are measured on your own trading: last year's margin spend against this year's, before and after, against the same reference. Nothing is projected, and nothing depends on which way the market moves.

The same outcome for a mid-sized company, equally illustrative: trading EUR 40m a year, 34 bps down to 11 bps is roughly EUR 92,000 a year. The arithmetic scales with volume; the measurement works the same at either size.

If the measurement instead shows your pricing is already fair, you hold independent, board-ready evidence of it. Just is a yearly subscription, paid upfront. If the savings we can document in your first year do not exceed what you paid us, the guarantee covers the difference.

Just vs your existing stack: market data, RFQ panel, TCA

Most treasury teams already run some of these tools: a market-data terminal, a multi-bank quote tool (an RFQ panel, where you ask several banks to quote and pick the best), and sometimes a cost report from a bank (transaction cost analysis, or TCA). Each answers a real question. One question stays open across all of them.

What it measuresReference it scores againstWho pays whomWhat it cannot tell you
Just The margin on every trade, in bps and money, and your position against a group of comparable companies. A real market rate: one you could actually have traded at, for your trade size and direction, at the moment of the trade. Shown on every scored trade. Paid by you, never the bank. No money from banks, no share of trading volume, no routing of trades. Which bank to trade with next. It measures pricing; it does not execute, route or advise.
Bank TCA Your trading cost over a period, usually reported on its own. Typically the interbank mid, a midpoint price nobody can trade on. Produced by or for the bank whose pricing it describes. Whether the pricing was fair. A mid-based number from the bank being measured carries no verdict the bank has to engage with.
Market data terminal Live market levels: mids and indicative spreads across currency pairs and settlement dates. The mid and rough quotes. Not tradable at your size or direction. You pay for the data; the quotes are contributed by banks and venues. What your own trades actually cost you, or what similar companies pay for comparable trading.
RFQ panel Which of the invited banks quoted best on each trade. The other quotes in the same auction, not the market itself. Typically paid for on the bank side of each trade. Whether any of the quotes was fair. Every bank on the panel can quote a wide margin and one still wins.

Each row describes the tool category, not any named vendor.

The quote tool and the terminal remain worth having; competition and market context answer their own questions. The benchmark answers the one they leave open: it measures the price you actually paid against a rate you could actually have traded at, and against what comparable companies pay. Industry guidance points the same way: bank spread grids (a bank's published margin table) are only indications, and real, observed trade data is the ground truth worth measuring against.

On the regulatory side, the FX Global Code (a voluntary standard most major FX banks have signed) commits the banks that sign it to be transparent about their mark-up. Transparency on request is useful; an independent measurement of every trade is a different level of assurance.

What it is not

It is not advice, and it is not a broker stepping between you and your banks. Just never contacts your banks, never touches your trades, and never takes a share of your trading. You keep the relationship direct; the benchmark replaces guesswork with evidence. On data: nothing you send is shared with another company, everything feeding the benchmark is pooled and stripped of identifying details first, and identifiable data is deleted when you leave.

FAQ

Is FX cost benchmarking the same as TCA?
No. TCA (transaction cost analysis) usually measures cost on its own against the interbank mid, a midpoint price nobody can trade on. Independent benchmarking scores each trade against the real market rate at the moment of the trade and against a group of similar companies, so the number carries a verdict.
What reference is each trade scored against?
A proprietary blend of independent market-data feeds, live bank quotes and several electronic trading venues (ECNs). It is deep-book, meaning tradable at your actual trade size and direction, and the reference rate is shown on every scored trade.
How are savings calculated?
Last year's margin spend minus this year's, on your own trading, before and after, against the same reference. Nothing is projected.
Do you contact our banks?
Never. The evidence is prepared for you; the conversation with the bank stays yours.
Who pays Just?
You do. Paid by you, never the bank, so our only incentive is your result.
Every trade scored against the real market rate, never the mid. Read the methodology
Terms used here: real market rate · basis point · similar companies · reference rate
Independent FX cost benchmarking for companies. Paid by you. Never by the bank.