Benchmark
Forex IB commission benchmarks 2026
Every partner programme in this industry quotes a headline percentage, and almost none of them quote it against the same base. Before a benchmark means anything, two offers have to be put in the same units.
Why two headline rates are not comparable
A revenue share is a fraction, and a fraction has a numerator and a denominator. The numerator is the part every programme advertises. The denominator is the part that decides what the number is worth, and it varies enormously between programmes that look identical in a comparison table.
A share of the spread a trader pays, a share of what a broker books as net revenue, and a share of what a partner network receives from the broker are three different bases. The same partner payment can be quoted as a small percentage of the first or a large percentage of the third. Neither figure is dishonest on its own. Putting them side by side without saying which is which is.
The five questions that normalise any offer
Ask these of every programme you are comparing, including ours, and write the answers in a row. The comparison becomes obvious once the row is complete.
Step 1 of 5. A percentage of what, exactly?
Get the base named. Spread paid by the trader, broker net revenue, or what the network receives. If the answer is vague, treat the rate as unknown rather than as generous.
Step 2 of 5. What event triggers a payment?
Signup, verification, first deposit, or every closed trade. A payment triggered by trading continues. A payment triggered by an account opening does not.
Step 3 of 5. What is clawed back, when, and on what event?
A clawback window on a front-end payment is normal. A clawback on a continuing revenue share is not, and it converts a headline rate into a rate with a discount you cannot forecast.
Step 4 of 5. Which trades do not count?
Minimum holding duration, abnormal spread conditions, hedged positions, particular instruments. Get the list in writing. This is where two identical rates become different money.
Step 5 of 5. How long does attribution last, and how is it captured?
A programme that attributes on a browser cookie loses conversions silently. A programme that attributes server side does not. Then ask how long the attribution survives after the match: lifetime, or a window.
Our own answers to those five
Published in full, so this page can be used against us.
- A percentage of whatThe same base at every rate in the band
- what Introbroker receives on the account
- The band
- 50% to 85%
- What triggers a payment
- Lifetime. It does not expire while the trader keeps trading.
- Clawback
- None
- If you take the hybrid instead
- 30 days on the front end
- Payout schedule
- Weekly, Monday 08:00 CEST
- Payout minimum
- $100
- How the rate movesRevenue received, never volume
- Reviewed on a rolling 30 days of revenue received
The benchmark table
When it is published it will be our own realised figures rather than a market survey, split by instrument class and by broker, refreshed quarterly, and stated with the denominator named on the same line. A benchmark built from other people's marketing pages is a summary of what the industry advertises, which is not the same thing as what it pays.
What to ignore in a commission comparison
- A ceiling rate with no stated route to it. If nobody will tell you what moves a partner up, the ceiling is a marketing figure.
- A rate quoted with no denominator, however large it is.
- Tier thresholds denominated in volume rather than in revenue. Volume is not a consistent unit across instruments, so two partners producing very different revenue climb the same ladder at the same speed.
- Any comparison chart with no churn control. Churn is the input that decides which model wins, and hiding it produces whichever answer the author wanted.
- A headline that compares a competitor's entry rate against the author's own ceiling.
Questions
Why does this page not contain a benchmark table?
Because we do not have figures we can stand behind yet. Two of our own broker reports imply different shares of the spread column and one of them implies a per-trade revenue below the cheapest instrument we quote, so the underlying economics are being resolved with the brokers before anything is published. Publishing a plausible range in the meantime would be inventing a figure on a page about money.
What will the benchmark contain when it is published?
Our own realised figures from our own ledger, split by instrument class and by broker, with the denominator named on the same line, refreshed quarterly. Not a survey of other programmes' advertised rates.
How do I compare offers in the meantime?
Use the five questions above. Get every programme's answer in writing, put them in one row each, and compare rows rather than headlines. In practice the denominator and the trade filters move the answer more than the headline rate does.
Where this comes from
Sources
Primary sources only: the regulator, the standards body, the platform vendor, the legislation, or our own agreements. Not another affiliate's summary of one.
- Our own documentIntrobroker, from each broker
Broker revenue statements and rebate reportsnot published
The per-broker figures the commission ledger reconciles against. Two of these do not currently reconcile with each other, which is why no page on this site publishes a per-lot figure.
- Our own documentIntrobroker
Introbroker partner agreementnot published
The commercial terms on this page in their contractual form: the rate, the denominator, the clawback position and the payout schedule. Sent before you sign.
- RegulatorFinancial Conduct Authority
Conduct of Business Sourcebook, COBS
What a firm communicating a financial promotion has to say, and how an inducement paid to an introducer is treated.