Where this came from
The desk began with a filing problem. Complaints about trading platforms arrive in a form that is almost impossible to publish: a long account of a bad experience, written by someone still inside it, with the details that matter buried under the details that hurt. Most sites solve that by turning the account into a star rating. We thought the rating was the least honest part, so we removed it and kept the filing.
What remains is dull by design. A file per platform. Reader correspondence in one column, the operator's answer in the other, and a written record of what we searched for and did not find.
Who we are
Three people produce everything here. Serafina Bozzano runs the complaints file and decides what in a reader's letter is evidence and what is distress. Marcus Threlfall handles the right of reply: he finds a contact of record where one exists, puts the substance of the complaint to it, and prints what comes back. Delia Achterberg works the records desk, checking names against public registers and writing the section of each report that lists what could not be established.
None of us claims credentials we cannot show you. We are editors, not regulators, lawyers or financial advisers, and the work here is journalism rather than supervision.
What we refuse to do
We do not rank. A ranking implies a comparison we are not equipped to make, and the brands covered here share too little verifiable information to be compared on anything but their advertising.
We do not publish testimonials. An untraceable quote of praise and an untraceable quote of complaint are the same object, and neither belongs on a page that asks to be trusted.
We do not fill gaps. If a report cannot say who operates a service, it says that, in those words, rather than reaching for a plausible company name.
We do not tell anyone to open or avoid an account. The decision belongs to the reader and to whoever advises him professionally.
The correction rule
Anyone can correct us: a reader, an operator, a regulator, a passing stranger with a document. A correction that changes the substance of a report is published as a correction, dated, with the previous claim named. Silent editing would make every earlier version of every page unreliable, which is a high price for looking as if we are never wrong.