The Honest State of Forex Affiliate Comparison Sites in 2026

If you have searched for a forex broker in the last five years, you have probably landed on a comparison site that ranked brokers in an order that mysteriously matched the size of each broker's affiliate payout. This piece breaks down how the industry actually works and what to demand from any comparison site before you trust its rankings.

How most affiliate comparison sites actually work

The honest description of the business model:

1. A broker pays the comparison site between USD 200 and USD 1,500 per "qualified" deposit (varies by broker, region, and account type). 2. The comparison site ranks brokers in an order that maximises its expected revenue per visitor. 3. The "methodology" page exists primarily to provide cover for that order.

This is not a moral failure unique to forex. It is the same revenue-driven sort that ranks insurance providers, web hosts, and credit cards across most of the affiliate web. The forex variant has two specific aggravating factors:

**Disclosure quality is uniformly terrible.** EU directives require affiliate disclosures, but most sites bury them in the footer in 8-point grey type. Almost none disclose the per-broker commission structure that determines the ranking.

**The product itself is dangerous.** 74-89% of retail CFD traders lose money. This is not a marketing slogan -- it is the mandatory disclosure number from each broker's KID (Key Information Document) under MiFID II. Routing a beginner to whichever broker pays the highest CPA can cause real financial harm.

The signals that distinguish editorial-led sites from pay-to-rank

When you arrive at a forex comparison site, run these five tests before trusting any ranking:

1. Is the methodology actually a methodology?

A real methodology specifies the weighting of each scoring category, the data source for each metric, and the review cadence. If it says "we evaluate brokers on safety, costs, platforms, and support" without specifying how, you are reading marketing copy. A useful test: pick a broker the site ranks number one. Ask whether the methodology, as stated, mathematically supports that ranking over the broker ranked third or fourth. If the answer is "you would have to ask the editors", the methodology is decorative.

2. Does the site ever criticise a broker by name?

This is the single fastest test. Visit the site's "best of" lists and look for negative comparisons by name. "Broker X has narrower spreads but charges higher overnight fees" is acceptable. "Some brokers have wider spreads" is not. Pay-to-rank sites avoid criticising any broker by name because every broker is a potential paying partner. Editorial sites name names.

3. Is there a "brokers we refused" surface?

A site that takes editorial standards seriously will publish a list of brokers it declined to cover, with reasons. Forex is a sector with active unauthorised firms, leveraged-scam operators, and brokers under enforcement action. A comparison site that has been operating for two years with zero refused brokers is either negligent or commercially captured.

4. Are paid placements visually and structurally separate?

Look for clearly labelled sponsored sections that sit apart from the editorial rankings -- a "Partner Brokers" carousel that never appears in the top-10 list, sponsored content filed under a permanent /sponsored-content/ path with a disclosure bar across the page, or paid placements rotated by URL parameter rather than embedded in the prose. If sponsored content is indistinguishable from editorial in layout or tone, the wall is decorative.

5. Does the site verify licences on the regulator's own register?

For every broker the site covers, the claimed regulator licence number should be cross-referenced against the regulator's own public register -- not against the broker's own claim page. CySEC, FCA, BaFin, MAS, SFC, and ASIC all maintain public registers that take 30 seconds to check. Comparison sites that do not link directly to the regulator entry for each broker are doing zero verification work themselves. The broker's footer is not a primary source.

What we do differently — and what it costs us

We launched fx-brokers.eu in 2026 specifically because we wanted to read the comparison site we could not find. The structural choices that distinguish us are not unique -- nothing on this page is technically novel. The difference is that we apply them at the cost of partner revenue we could otherwise capture.

Editorial scoring is published as methodology

Our [methodology](/methodology) lists the weighting for each scoring category and the cadence on which we re-score every broker. The composite score is a deterministic function of the inputs, not an editor's preference. When we update an input -- a broker's published spread, a regulator enforcement action, a withdrawal complaint pattern -- the score moves, and we publish the lastVerified date alongside.

This costs us roughly 15% of the revenue we could earn from juicing rankings. We accept that cost.

Paid placements sit structurally separate

Our rate card is on request, not on the site (per the operator's standing instruction). When a broker becomes a paid partner, they appear in a clearly labelled "Partner Brokers" carousel — never in the editorial top-10 list. Sponsored content lives at /sponsored-content/ with a permanent disclosure bar. We have refused two requests to "elevate" a broker into editorial positions in exchange for higher CPAs. Both refusals are documented on the [/editorial-standards/brokers-we-refused](/editorial-standards/brokers-we-refused) page.

This costs us roughly 25% of the revenue we could earn from blurring the boundary. We accept that cost.

Every licence number is verified on the regulator's own register

Each broker review shows a "verified at" date and a one-click link to the regulator's public register entry. The verification data is in our codebase at /data/broker-verifications.ts and is public. When a broker's licence is withdrawn, restricted, or transferred to a different entity, we update the review within seven days.

This costs us roughly 8% of the revenue from brokers whose licence status has degraded but who still pay -- we list them honestly even when it depresses their conversion. We accept that cost.

Brokers we refused is public

Some brokers we declined to cover are listed on [/editorial-standards/brokers-we-refused](/editorial-standards/brokers-we-refused) with the specific reason. Reasons range from "licence inconsistencies on the regulator register" to "unresolvable client complaint pattern" to "active enforcement action". The page is updated as new refusals occur. This is the most uncomfortable part of the site for our partner brokers. We keep it public anyway.

What you should actually do

If you are choosing a forex broker right now, ignore the rankings on any comparison site (including this one) until you have done your own three-minute verification:

1. **Find the claimed regulator licence number in the broker's website footer.** 2. **Open the regulator's public register** (CySEC, FCA, BaFin, MAS, SFC, ASIC) and search for the broker by name. 3. **Confirm the entity name and licence status match exactly.** Confirm the licence is "Active" or "Authorised", not "Withdrawn" or "Restricted". 4. **Search the regulator's warning list and enforcement database** for the broker's name. 5. **Read the broker's KID document** -- it contains the mandatory disclosure of retail-account loss rates, the spreads on the major pairs, and the leverage caps applicable to your jurisdiction.

If a comparison site recommends a broker that fails any of these checks, the comparison site is not doing the work.

What we would change in the industry

Three structural changes would improve the affiliate-comparison sector materially:

**Mandatory CPA disclosure.** Comparison sites should be required to publish the commission they receive per partner. Not a generic "we may earn commission" line in the footer -- the specific dollar amount per broker. This single change would eliminate 90% of pay-to-rank distortion overnight, because the reader would instantly see why a particular broker is recommended.

**Regulator-side broker registries with machine-readable APIs.** CySEC, FCA, MAS, and SFC publish their broker registers as human-readable HTML. None offer machine-readable APIs. A comparison site that wanted to keep licence data fresh has to scrape the regulator's own site, which is fragile. An API would make automated weekly verification trivial.

**Independent audit.** Affiliate-comparison sites could submit to an annual editorial audit by a body like the IAB or ASA, with the audit results published. The IAB has frameworks for digital advertising disclosure that translate cleanly to comparison content. There is no equivalent for affiliate-led comparison rankings.

None of these will happen in 2026. The status quo serves the larger comparison sites and the larger brokers. The cost is borne by retail clients who route to whichever broker the comparison site's revenue model favours.

Conclusion

The forex affiliate-comparison sector in 2026 is structurally captured. Most comparison sites have business models that reward ranking distortion. The five tests above let you sort the editorial-led sites from the pay-to-rank sites in under three minutes per site.

We built fx-brokers.eu to be readable for someone who is going to run those five tests against us. If we fail any of them, the criticism should be public. If we pass them, the rankings are worth more.

*This article reflects the operator's editorial standards as of May 2026. The full /editorial-standards path on fx-brokers.eu documents what we will and will not do under partner pressure.*