◉The Seal Desk Open the partner account
paid
Affiliate disclosure. The partner link in the masthead and in the bands beside the copy on this page is a sponsored link to a partner operator, and this site may be paid if you open an account through it, at no extra cost to you. It carries rel="sponsored noopener" and opens in a new tab. That matters on this desk in particular: the subject is how much evidence a third-party mark carries, and this site carries commercial marks of its own. No operator, testing house, review platform, award or association is named, rated or recommended anywhere on this site.
The Seal Desk / The marks
Desk 64 - the marks that vouch

Six beliefs, judged against the arithmetic

The beliefs below are the ones a page is designed to produce. Each is checked against a figure from the samples rather than against an opinion, and the verdict is about the belief as stated, not about any issuer.

Mark stub
Verdicts
6
False
2
Partly
3
Holds up
1
the issuerThe company or body that put the mark there - a testing house, an association, a review platform or a media brand
the scopeWhat the examination covered: a process, a system, a game type, one entity or one market
the recourseWhat the issuer does when the operator fails - the question almost no mark answers
Direct answerSix beliefs are checked here. Two are false, three are partly true and one holds up: a mark whose issuer examined the site, whose scope is published, whose entity matches your terms and whose issuer ends it on a failure is genuine evidence - and that combination is a minority of the badges on a page.

The six

False

“More badges means a safer site”

Eleven marks on the sample page, of which three were examined and five are payment and helpline logos. The count measures how much a page displays, and display is cheap: a bought mark costs 2,400.00 a year against 6,400.00 for one examined process.

Partly

“A seal means the site was audited”

True only if the scope says what was audited. Sample 3’s examination covers one of thirty-seven documented processes, 2.7% of the estate, and none of the twelve documents in the file is published. An audit happened; what it covered is the question.

Partly

“A good star rating means customers are happy”

A mean of 4.4 over 1,940 reviews is equally consistent with a spread of fours and fives and with 85% five-star and 15% one-star. The number survives the loss of the shape, and the shape is where the customers with a serious complaint live.

False

“An award means the operator is the best”

An award is a decision on a calendar, held for a year, in a category the publisher defines, sometimes entered for a fee of 2,400.00. It is a true statement about one year’s decision and it cannot lapse when the operator’s conduct changes.

Partly

“A seal on the page covers this site”

A certificate names a legal entity. Three of the eleven marks in sample 4 name a group company that is not the one on the terms page, and the licence number displayed belongs to an entity holding two of the group’s five brands.

Holds up

“A mark you can look up is worth looking up”

Where an issuer names itself, publishes a scope, keeps a register and ends the mark on a failure, a reader can check the certificate, match the entity and reach a finding. That is 31 of the 84 checkable marks in the book, and it is the only kind of mark worth the six minutes.

How the verdicts were reached

Each belief was rewritten as a claim that could be false, and then checked against a figure already on this desk. None of the verdicts depends on a real operator’s conduct, and none of them says that any particular issuer is dishonest - the samples are invented and the arithmetic is the only evidence in play.

What would change a verdict. A published scope, a register entry for the entity on the reader’s terms, a rating shown with its distribution, and an issuer that ends a mark on a failure would each move a “partly” toward “holds up”. The verdicts are about the marks as they are displayed, not about what the issuers could publish.