Most comparison tables are built out of what operators choose to say. This page is built out of what they do not, because in a library of 100 brands the empty cells outnumber the full ones in every category except one.
The terms and register entries behind these counts were read on 26 August 2026, with one brand re-read on 2 September 2026. Every figure below is a count of fields that came back empty, and the second half of the page explains why an empty field is a gap rather than a finding.
How much is missing, field by field?
| Field | Brands that publish it | Brands that do not |
|---|---|---|
| Identity-verification clause | 49 | 51 |
| Licence number | 68 | 32 |
| Restricted-country list | 38 | 62 |
| Supported-coin list | 30 | 70 |
| Withdrawal ceiling with a clause number | 14 | 86 |
| Year the business was founded | 11 | 89 |
| A money figure in the verification clause | 5 | 95 |
The licence number is the only field a clear majority publish. Everything else is a minority disclosure, and two of the rows — the founding year and the verification threshold — are published by roughly one operator in ten or fewer.
Why can so little be checked at all?
Because the documents themselves are not reliably reachable. Of the hundred brands, 38 had their pages read live, 41 could only be read from an archived copy of the same pages, and 21 returned nothing usable from either route.
Those 21 are the hardest number on this page. For them every single field is empty, not because the operator publishes nothing but because nothing could be read. A further group of brands could not be reached at all when the library was assembled, and no review of them appears anywhere on this site for that reason.
Twenty-one blank rows in a hundred is not a rounding error in a comparison. It is a fifth of the field about which the honest answer is “unknown”.
What does an empty field actually mean?
One of four things, and the four are not interchangeable:
- The operator publishes the fact and our reading did not reach it — behind a login, inside a downloadable file, or on a page that failed to load.
- The operator publishes a document that genuinely does not contain that clause.
- The operator’s pages could not be read at all, so the field was never in play.
- The fact exists but is not a published fact: an internal limit, an unpublished policy, a rule applied case by case.
Only the second of those is a statement about the operator. The other three are statements about the reading.
Where does this go wrong in practice?
In the two places where an absence reads as good news, and neither of them survives contact with the documents.
The first is the withdrawal ceiling. Eighty-six brands publish no cap. A table that renders that as “no limit” has invented a term of contract. The 14 that do publish one put it in writing with a clause number, and their figures range from 4,000 euro a day to 1,000,000 USDT a week — a spread wide enough that guessing what an unpublished cap might be is meaningless.
The second is the verification clause. Forty-nine clauses were read and all 49 leave the decision to request documents with the operator; only five name a money figure at all. Reading the 51 silent brands as having no verification requirement inverts what the read clauses actually say.
An unpublished rule is not a lenient rule. It is an unpublished rule.
Why does a scoring system have to count this?
Because a score computed only over the fields an operator filled in rewards operators for publishing less. A brand that discloses nothing has no bad fields; a brand that publishes a modest withdrawal cap and an explicit verification threshold has two fields that can be marked down.
Any comparison that divides by the number of fields present rather than by the number of fields possible will therefore float the least transparent operators to the top. The correct denominator is every field the library asks for, and a missing field costs the brand rather than excusing it.
The same discipline applies to registers. Fifty-seven of the hundred domains returned no Curaçao certificate, but 39 appear in the Anjouan register instead, so “absent from one register” and “unlicensed” are different statements. The two documents are described on our pages about the Curaçao certificate and the Anjouan register.
Which gaps matter most to a reader?
- The verification clause, because it governs the moment money tries to leave. Where it is unread, nothing at all is known about what triggers a document request.
- The withdrawal ceiling, because it determines how long a large balance takes to come back, and 86 brands say nothing about it.
- The restricted-country list, because it is the operator’s own written position on a country, and 62 brands leave it blank.
- The licence status, because it is the only field in the whole library that changes without any notice to anyone.
Where India specifically appears in the third of those, and what the clause says, is on our page about operators that restrict India in their terms.
What this site does with a gap
Leaves it visible. A field with no source is recorded as unknown and printed as unknown, with the date of the attempt. It is never filled in from a competitor’s table, from another review site, or from an operator’s marketing copy, and it is never converted into a positive claim about the operator.
None of this bears on Indian law, which is a separate document with its own text; that text is set out on our page about the Online Gaming Act, 2025. A licence field, full or empty, answers nothing that statute asks.