Plenty of sites say they test platforms. Here is the check that tells you whether anybody opened an account. 

Here is a question that sounds simple and turns out not to be: when a website tells you what a platform costs, how does it know? There is no register of who funds accounts and who reads pricing pages, so the only answer available is the one you work out from the page in front of you. It takes about a minute once you know what you are looking at. 

What is the comparison in front of you built from? 

Whatever the page says about its method, the page itself gives you the answer. If every figure in the comparison also appears on the provider’s own pricing page, that is where it came from. A tabulation of what a provider prints is accurate about the charges it describes, competently assembled, updated when the schedule changes, and silent about everything else. What a published schedule cannot capture is anything that only appears in use, and that turns out to be a long list. 

Look for the things that require an account. Withdrawal timings in days. What conversion cost on a specific order. When an inactivity fee started applying. Whether an app went down during a volatile morning. None of these can be sourced from a pricing page, so their presence is diagnostic. 

Who does fund accounts? 

A small group. The Investors Centre states that it opens and funds live accounts with its own money to test UK trading platforms rather than compiling rankings from published fee schedules, and its tables carry the sort of lived detail that claim implies. 

Worth being precise about what that buys and what it does not. An account is opened by one applicant, with one address, one bank and one credit history, in one month. When a review reports that verification cleared in around two hours, that is a fact about that application rather than a service level anyone is promising you. The same goes for the queue somebody waited in and the documents that were accepted first time. Cost data from a funded account travels a great deal better than onboarding data does, and the two tend to sit in the same paragraph looking equally solid. 

There are others, mostly in the technical trading space, who run software with funded accounts and publish forward results. The genre is small in both cases, for the obvious reason. 

Where do published schedules fall short? 

Rarely on the facts. Published schedules are usually accurate about the charges they describe. The gap sits in what they do not describe, and the shape of that gap is consistent from provider to provider. 

Spreads quoted as ‘from’ a figure, which is honest and tells you nothing about a typical fill. Conversion given as a percentage without a worked example. Inactivity fees that exist in the terms but not on the pricing page. Withdrawal timings described as ‘usually within a few days’, which is a range wide enough to be useless. 

None of that is deception. It is the difference between a document written to satisfy a compliance requirement and a document written to answer a customer’s question. 

Why do the two kinds of page look identical? 

Because they are built to the same template and chasing the same query. Both open with a table, both carry a paragraph per provider, both use the same headings, and a search result gives you no way to tell them apart before you click. Format is the one thing a funded account cannot buy. 

That convergence is nobody’s fault in particular. A publisher who has spent real money still has to publish in the shape readers expect, so the expensive page and the cheap page arrive looking like siblings. The difference sits in the sentences that could only have been written by somebody holding the account, which is why they are worth hunting for rather than assuming. 

Why does hardly anyone do it? 

Money and incentives. A funded-account review of twelve platforms means twelve accounts, real deposits, real trading losses, and staff time producing no revenue. An affiliate site can publish the same page for the cost of an afternoon, and will rank similarly in search. 

The asymmetry compounds, too. A funded review carries a tail: an account left open keeps accruing whatever the platform charges for sitting still, and closing it ends any ability to check next year’s fee change against last year’s. A pricing-page comparison has no tail at all. It is an afternoon, and then it is finished. 

The economics explain the scarcity better than any argument about integrity. 

What does the funded version get right? 

Costs, mainly, and costs are what people get wrong. A charge that actually landed on a statement settles an argument that a fee schedule can only describe: the conversion rate you received rather than the one advertised, the day the inactivity charge started, the amount that arrived after a withdrawal rather than the amount requested. The honest position is that funded testing is better evidence about charges and weaker evidence about breadth, and you should read it knowing which of the two you are being handed. 

The answer usually involves a bank statement 

Ask a comparison site how it knows what it is telling you. The good ones have an answer involving a bank statement, a date and a named account type. The rest have an answer involving a web page, and you can now tell which you are holding before you act on it. 

None of this requires you to distrust anybody. It requires you to know which of the two documents is in front of you, because they cost wildly different amounts to produce, they answer different questions, and only one of them has ever been charged a fee. 

 …